{ "schema_version": 1, "status": "Inventory of what the v1 memory and workspace could become in v2. Every row is a proposal; agent-catalog.json and frequent-operations.json remain the authoritative definitions of what is designed.", "census": { "total": 119, "by_kind": { "skill": 30, "tool": 68, "article": 20, "agent": 1 }, "by_status": { "existing": 39, "candidate": 60, "article": 20 }, "by_domain": { "support-root": 10, "data-transfers": 14, "data-pricing": 8, "data-printing": 6, "data-access": 9, "data-health": 7, "data-master": 8, "data-operations": 6, "configuration": 12, "source": 9, "commercial": 4, "common": 16, "workspace": 10 }, "seed_files_cited": 515 }, "capabilities": [ { "id": "support.skill.classification", "kind": "skill", "label": "Classify a request against the contract", "one_line": "Decide §1 Standard severity or §2 Personalization priority, detect a CR trigger and set the clocks.", "owner": "support.root", "domain": "support-root", "status": "existing", "existing_ref": "support.skill.classification", "seeds": [ ".claude/skills/classify-ticket/skill.md", "knowledge/support_contract_bg.md", "knowledge/support_procedures.md", "memory/reference/support_contract_custom_bg.md", "memory/reference/support_contract_custom_en.md", "memory/feedback/config_toggle_vs_cr_classification.md" ], "evidence": "A classification block was written on 322 of the 332 analysed tickets of the 02.09 run (§1 296 · §2 19).", "evidence_row": null }, { "id": "support.tool.sla-clock", "kind": "tool", "label": "SLA clock and breach calculator", "one_line": "Turn severity, contract section and calendar into response/resolution deadlines, applying the §1.5.3 workaround downgrade.", "owner": "support.root", "domain": "support-root", "status": "candidate", "existing_ref": null, "seeds": [ "knowledge/support_contract_bg.md", "knowledge/support_procedures.md", "memory/reference/support_contract_custom_en.md", ".claude/skills/classify-ticket/skill.md" ], "evidence": "Severity distribution over the same 322 classified analyses: Sev 3 151 · Sev 4 43 · Sev 2 16 · Sev 1 3.", "evidence_row": null }, { "id": "common.skill.communication", "kind": "skill", "label": "Prepare and deliver the customer outcome", "one_line": "Draft the Bulgarian customer reply and the internal note, then deliver them through the gated send.", "owner": "common.communicator", "domain": "support-root", "status": "existing", "existing_ref": "common.skill.communication", "seeds": [ "memory/feedback/customer_reply_on_point.md", "memory/feedback/customer_reply_after_applied_fix.md", "memory/feedback/outbound_technical_reply_summary_then_details.md", "memory/feedback/ticket_closure_via_pending.md", "memory/feedback/aisa_internal_comment_single_per_ticket.md" ], "evidence": "The desk's skill #1: 330 tickets carry a customer conversation phase; 4 holders.", "evidence_row": 1 }, { "id": "support.skill.channel-access", "kind": "skill", "label": "Diagnose a ticket-portal access problem", "one_line": "Read-only sweep of the ticket system's account, groups, product access and portal reachability, ending in fix steps for an org admin.", "owner": "support.root", "domain": "support-root", "status": "existing", "existing_ref": "support.skill.channel-access", "seeds": [ ".claude/skills/jira-access/skill.md", "knowledge/atlassian_jira_admin.md", "memory/reference/jira_customer_portal_no_access.md", "memory/reference/jsm_portal_no_access_wrong_group.md", "memory/reference/chrome_devtools_mcp.md" ], "evidence": "3 memory/knowledge files plus one v1 skill; no independent ticket count was measured.", "evidence_row": null }, { "id": "support.tool.origin-dedup", "kind": "tool", "label": "Deduplicate one request across the three arrival channels", "one_line": "Match a ticket, an OTRS ticket and a mail thread onto one origin key and report which of them is the freshest.", "owner": "support.root", "domain": "support-root", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/cross_channel_correlation.md", "memory/feedback/duplicate_request_verify_prior_residuals.md", "memory/reference/hdesk_jira_provenance_staleness_check.md", "memory/feedback/scan_mailbox_and_microservice_logs_per_subject.md", "hdesk/scripts/gen_status.ps1" ], "evidence": "4 memory files plus the queue-status generator; the desk's dedup-on-origin-key skill has no catalogue entry.", "evidence_row": null }, { "id": "support.tool.hdesk-transfer", "kind": "tool", "label": "OTRS ticket to ServiceDesk transfer and refresh", "one_line": "Resolve the OTRS id, fetch the ticket, find or create its ServiceDesk counterpart and re-post only the new articles.", "owner": "support.root", "domain": "support-root", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/transfer-hdesk/skill.md", "hdesk/scripts/gen_status.ps1", "hdesk/scripts/gen_cached_views.ps1", "memory/reference/aisa_hdesk_cli.md", "memory/feedback/hdesk_refresh_show_summary_table.md", "memory/project/hdesk_pending_jira_updates_backlog.md" ], "evidence": "The most mechanical of the 20 v1 skills — end to end already CLI-driven; 39 of the 52 OTRS-queue configuration tickets arrive this way (D59).", "evidence_row": null }, { "id": "support.tool.pattern-match", "kind": "tool", "label": "Match a symptom against the known-pattern catalogue", "one_line": "Score a new request's symptom text against the recorded symptom-cause-fix patterns and the open known-issue register, and return ranked candidates with their evidence.", "owner": "support.root", "domain": "support-root", "status": "candidate", "existing_ref": null, "seeds": [ "knowledge/common_patterns.md", "KNOWN_ISSUES.md", "memory/feedback/maintain_known_issues_register.md", "scripts/triage_baseline.py", "memory/reference/kb_memory_vs_rediscovery_cost.md" ], "evidence": "56 open known issues and a symptom-cause-fix table; the v1 grep baseline for this retrieval is measured in Measurements § TB.", "evidence_row": null }, { "id": "support.skill.queue-sweep", "kind": "skill", "label": "Periodic re-read of the open case set", "one_line": "Walk the open cases, detect what changed since the last pass and raise the ones that need a decision.", "owner": "support.root", "domain": "support-root", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/sd-monitor/skill.md", "memory/feedback/monitor_jira.md", "memory/feedback/pre_post_status_refresh.md", "scripts/build_jira_summary.py" ], "evidence": "v2 replaces the polling half with platform.intake; the change-detection half of /sd-monitor has no catalogue entry.", "evidence_row": null }, { "id": "support.article.desk-practice", "kind": "article", "label": "Desk practice: statuses, transitions, tags, closure", "one_line": "How this desk moves a case through its statuses, who is named on it and how it is closed back to the customer.", "owner": "support", "domain": "support-root", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/ticket_closure_via_pending.md", "memory/feedback/monitor_jira.md", "memory/feedback/jira_via_aisa_jira_cli.md", "memory/feedback/sd_prefix_jira_only_not_sr_policy_id.md", "memory/reference/jira_adf_media_embedding.md", "memory/feedback/jira_adf_date_node_invisible_in_list.md", "knowledge/persons.md" ], "evidence": "7 memory/knowledge files describe the desk's own ticket handling.", "evidence_row": null }, { "id": "support.article.scope-and-cr", "kind": "article", "label": "Scope, change requests and what the contract excludes", "one_line": "When a request stops being covered support and becomes a change request, a personalization or someone else's system.", "owner": "support", "domain": "support-root", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/insis_change_request.md", "memory/feedback/config_toggle_vs_cr_classification.md", "memory/feedback/check_test_db_before_cr.md", "memory/feedback/find_the_requirement_before_proposing_change.md", "memory/feedback/external_regulators_no_action_offers.md", "memory/feedback/offticket_policy_changes_need_approval.md", "memory/feedback/recurring_manual_fix_escalate_to_config.md" ], "evidence": "7 feedback files; 22 of 485 tickets were classified CR candidate / new feature with a 0 % DB-write rate.", "evidence_row": null }, { "id": "data.skill.transfer-diagnosis", "kind": "skill", "label": "Diagnose transfers and families", "one_line": "Classify a stuck or wrong transfer into blocked validation, mapping, duplicate or zombie transaction and emit the bounded fix.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "existing", "existing_ref": "data.skill.transfer-diagnosis", "seeds": [ ".claude/skills/check-transfer/skill.md", "memory/reference/ipal_transfer_paths.md", "memory/reference/transfer_already_started_idempotency_guard.md", "knowledge/transfer_pipeline.md", "memory/feedback/prefer_return_to_application_for_dual_system_corrections.md" ], "evidence": "112 tickets asked / 160 analyses used; 171 → 108 h; 4 holders.", "evidence_row": 2 }, { "id": "helper.transfer-trace", "kind": "tool", "label": "Transfer trace: pipeline log, notes and the log window in one read", "one_line": "Read the migration log, the user notes and the environment's log stack for one policy or window and return the ordered failure chain.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "existing", "existing_ref": "helper.transfer-trace", "seeds": [ "memory/reference/kibana_log_access.md", "memory/reference/transfer_already_started_idempotency_guard.md", "knowledge/transfer_pipeline.md", "memory/reference/int_trnsf_pend_bulk_triage.md" ], "evidence": "Serves 6 operations in the inventory; the trace shape itself is 45 tickets asked / 45 analyses used.", "evidence_row": 4 }, { "id": "helper.policy-family-compare", "kind": "tool", "label": "Both-administration policy and family comparator", "one_line": "Compare one policy or one multi-year family field by field across both administrations and return the ordered differences.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "existing", "existing_ref": "helper.policy-family-compare", "seeds": [ "sql/ref/srd_integr_compare_ipal_insis_car.sql", "memory/reference/policy_sync_compare_checklist.md", "memory/reference/multiyear_policy_verification_checklist.md", "memory/reference/multiyear_policy.md", "memory/feedback/policy_sync_checklist_living.md" ], "evidence": "Serves 13 operations; the compare shape is 42 / 31 and the multi-year repair 25 / 39.", "evidence_row": 7 }, { "id": "data.skill.policy-compare", "kind": "skill", "label": "Compare one policy across both administrations", "one_line": "Run the living sync checklist over a policy and produce the ordered correction set with a post-fix difference of zero.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "existing", "existing_ref": "data.skill.policy-compare", "seeds": [ ".claude/skills/analyze-policy/skill.md", "memory/reference/policy_sync_compare_checklist.md", "memory/reference/insis_policy_states.md", "memory/reference/c_policies_status_semantics.md", "memory/reference/ipal_insis_status_desync.md" ], "evidence": "42 tickets asked / 31 analyses used; 62 → 43 h; 4 holders.", "evidence_row": 7 }, { "id": "data.skill.framework-sync", "kind": "skill", "label": "Synchronise a cargo framework for child issuance", "one_line": "Walk the framework's seven layers across both administrations and emit the ordered SQL that lets the child policy issue.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "existing", "existing_ref": "data.skill.framework-sync", "seeds": [ ".claude/skills/sync-framework/skill.md", "memory/reference/cargo_1101_family_sync.md", "memory/reference/cargo_framework_extension_workflow.md", "memory/reference/cargo_products.md", "memory/reference/cargo_1100_single_policy_sync_footprint.md", "memory/reference/average_agent_flow.md" ], "evidence": "Framework sync 13 / 9 and framework extension 14 / 13; 3 holders each.", "evidence_row": 20 }, { "id": "data.tool.id-resolver", "kind": "tool", "label": "Resolve any estate identifier to its object", "one_line": "Take a 12-digit id, policy number or mapping id and return what it is and its counterpart in the other administration.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/decode_id_before_assuming_policy_no.md", "memory/reference/ipal_insis_id_mapping_tables.md", "memory/reference/policy_no_eq_sr_policy_id_post_transfer.md", "memory/reference/policy_no_composition.md", "memory/reference/ipal_id_sequence_prefix_convention.md", "memory/reference/source_id_vs_client_id_distinction.md", "memory/reference/insis_ipal_object_mapping.md", "memory/reference/crossed_insis_mapping_without_chain_contamination.md", "memory/reference/policy_no_extended_fleet_format.md" ], "evidence": "9 memory files; 'id decoding' is one of the 13 rows of the inherited v1 methodology catalog.", "evidence_row": null }, { "id": "data.tool.chain-walk", "kind": "tool", "label": "Object and annex version-chain dump", "one_line": "Dump a whole object or annex version chain with its counterpart pairing and policy bindings as one table.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_chain_walk_connect_by.md", "memory/reference/sr_objects_chain_semantics.md", "memory/reference/parallel_sr_objects_trees.md", "memory/reference/ipal_annex_model.md", "memory/reference/annex_transfer.md", "sql/proposed/sd1599_reconcile_obj_car.sql" ], "evidence": "6 memory/SQL seeds; the query template is already written out verbatim in memory.", "evidence_row": null }, { "id": "data.tool.participant-compare", "kind": "tool", "label": "Participant comparator across both administrations", "one_line": "Compare owner, lessee, beneficiary and policyholder rows for one policy and return the missing propagation.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/insis_object_participant_tables.md", "memory/reference/leasing_policy_data_model.md", "memory/reference/pholder_swap_ipal_only_propagation_gap.md", "memory/reference/ipal_non_vehicle_object_owner_transfer_gap.md", "memory/reference/object_credited_partial_amount_cr.md", "memory/feedback/verify_participant_org_from_live_header.md", "sql/proposed/SD-1967_owner_correction_insis.sql", "sql/proposed/SD-1997_fix_active_4800_train_owners.sql" ], "evidence": "Participant repair 31 / 25 and policyholder swap 10 / 26 (the swap shape is thin: 2 holders).", "evidence_row": 9 }, { "id": "data.tool.history-audit-walk", "kind": "tool", "label": "History-table walk with writer attribution", "one_line": "Read the audit history of one object over a window and attribute every change to the account and transaction that made it.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/hist_o_car_audit_pattern.md", "memory/reference/hist_p_people_audit.md", "memory/feedback/short_name_check_hist_first.md", "memory/project/serdica_access_cross_row_mutation_open.md", "memory/reference/abc_polpr_log.md" ], "evidence": "5 memory files; the cross-row mutation signature it detects is an open project item.", "evidence_row": null }, { "id": "data.tool.cancel-propagation-check", "kind": "tool", "label": "Cancellation propagation check", "one_line": "Confirm a cancellation or termination reached the counterpart administration and the regulator, and name the layer that did not move.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_cancel_confirm_double_transfer_gap.md", "memory/reference/ipal_cancel_transfer_timeout_insis_live.md", "memory/reference/ipal_cancel_annex_begin_date_default.md", "memory/reference/cancelled_policy_restoration_paths.md", "memory/reference/gc_period_insis_gf_desync.md", "sql/proposed/301001168834_01_finish_cancellation_from_ipal.sql" ], "evidence": "Cancel not propagated 16 / 74 and restore after mistaken cancel 12 / 25.", "evidence_row": 15 }, { "id": "data.tool.sync-policy-enddate", "kind": "tool", "label": "Policy end-date and duration sync", "one_line": "Push a corrected policy end date and its duration dimension to the counterpart administration for one policy number.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "sql/proposed/SD-1998_sync_cpolicy_to_insis_enddate_duration.sql", "sql/ref/sync_cpolicy_to_insis_enddate_prod_20260730.sql", "memory/reference/insis_period_desync_convert_application.md", "memory/reference/ipal_policy_period_month_end_clamp.md" ], "evidence": "A deployable parameterised procedure already exists in the checkout, awaiting promotion; period reconciliation is one of the inventory's transfer operations.", "evidence_row": null }, { "id": "data.skill.beth-triage", "kind": "skill", "label": "Triage an assistant-driven action failure", "one_line": "Separate an outage, a validation refusal and a data defect behind a failed assistant action, and reconcile whatever it half-applied.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/beth_assistant.md", "memory/reference/beth_chat_assistant.md", "memory/reference/beth_processbatch_distributed_tx_blocking.md", "memory/reference/beth_change_vehicle_application_gap.md", "memory/reference/beth_changeown_manual_prem_zero.md", "memory/reference/beth_vehicle_object_chain_loop.md", "memory/reference/bethgpt_changeown_workflow.md", "memory/reference/beth_annex_print_regen.md" ], "evidence": "14 tickets asked / 24 analyses used; 18 → 12 h; 3 holders; 8 memory files. No catalogue entry.", "evidence_row": 19 }, { "id": "data.tool.process-instance-probe", "kind": "tool", "label": "Process definition and instance probe", "one_line": "List the process definitions registered for a product and the live instances for one policy, and classify each as running, halted or completed late.", "owner": "support.investigator.transfers", "domain": "data-transfers", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_camunda_midflow_halt.md", "memory/reference/camunda_engine_getusertasks_persistence_400.md", "memory/reference/core_camunda_in_process_bottleneck.md", "memory/reference/bpmn_endevent_id_to_handler_switch.md", "memory/reference/ipal_completed_process_late_completion.md" ], "evidence": "1 ticket asked / 34 analyses used — the widest gap between asked and used in the catalogue; 3 holders.", "evidence_row": 35 }, { "id": "data.skill.fresh-investigation", "kind": "skill", "label": "Investigate from fresh evidence", "one_line": "Establish the mechanism from current observations when no experience matches, and return an explicit unknown with the next separating evidence.", "owner": "support.investigator", "domain": "data-transfers", "status": "existing", "existing_ref": "data.skill.fresh-investigation", "seeds": [ ".claude/skills/investigate-ticket/skill.md", "memory/feedback/verify_dont_guess.md", "memory/feedback/cross_check_and_reanalysis.md", "memory/feedback/db_investigation_discipline.md", "memory/feedback/failed_probe_is_not_a_finding.md" ], "evidence": "81 of 485 tickets are incident/defect, the class with no matching prior shape.", "evidence_row": null }, { "id": "data.skill.pricing-diagnosis", "kind": "skill", "label": "Diagnose pricing and installments", "one_line": "Explain or correct a premium component, a discount, a loading or an installment plan on an issued policy.", "owner": "support.investigator.pricing", "domain": "data-pricing", "status": "existing", "existing_ref": "data.skill.pricing-diagnosis", "seeds": [ "memory/feedback/premium_ld_applied_vs_fed_factor.md", "memory/reference/insprem_premium_spec.md", "memory/reference/billing_and_installments.md", "memory/reference/casco_discount_cap_pfrdisc.md" ], "evidence": "Discount/loading/deductible fix 40 / 77 and installment plan 30 / 75; 4 holders each.", "evidence_row": 5 }, { "id": "helper.rating-replay", "kind": "tool", "label": "Rating request-versus-result comparator", "one_line": "Replay the rating call for a selected cover set and compare base premium, every loading and tax row and the roll-up against the stored result.", "owner": "configuration.abacus.validator", "domain": "data-pricing", "status": "existing", "existing_ref": "helper.rating-replay", "seeds": [ "memory/reference/abacus_getrates_direct_call.md", "memory/feedback/abacus_rates_presentation.md", "memory/feedback/premium_ld_applied_vs_fed_factor.md", "memory/reference/abacus_validation_pipeline.md" ], "evidence": "27 tickets asked / 62 analyses used; shared with the configuration stage test; 4 holders.", "evidence_row": 12 }, { "id": "configuration.tool.getrates-client", "kind": "tool", "label": "Rating-engine call builder and response normaliser", "one_line": "Build the rating request from what the product declares, call the engine and normalise the offers catalogue it returns.", "owner": "configuration.abacus.builder", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/abacus_getrates_direct_call.md", "memory/reference/getrates_request_builder_ownership.md", "memory/feedback/getrates_response_is_offers_catalog.md", "memory/reference/graphql_prpricingfactorvalues.md", "memory/reference/abc_cfg_pricing_factors_mapping.md" ], "evidence": "5 memory files; the engine computes and stores nothing, so the call itself is the safe inner loop of the stage test.", "evidence_row": null }, { "id": "data.tool.premium-reconcile", "kind": "tool", "label": "Stored-premium reconciliation across both administrations", "one_line": "Compare the per-cover premium, loading and discount rows of the two administrations for one policy and return the reconciling difference.", "owner": "support.investigator.pricing", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/premium_ld_applied_vs_fed_factor.md", "memory/reference/insprem_premium_spec.md", "memory/reference/declared_but_unrated_ld_pattern.md", "memory/feedback/pol_prem_rate_manually_corrected_ambiguity.md", "memory/feedback/systemic_premium_scan_hygiene.md", "memory/reference/abacus_ipal_ld_catalogue_mismatch.md", "memory/reference/ppa_bpshift_manual_premium.md" ], "evidence": "40 tickets asked / 77 analyses used; 64 → 44 h; distinct from the rating replay, which questions the engine rather than the stored rows.", "evidence_row": 5 }, { "id": "data.tool.installment-plan-recompute", "kind": "tool", "label": "Installment plan recompute and comparator", "one_line": "Recompute the installment plan from the premium and frequency, compare both administrations' plans and emit the correction that keeps paid instalments untouched.", "owner": "support.investigator.pricing", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/change_installments_procedure.md", "memory/reference/billing_and_installments.md", "memory/reference/installment_rounding.md", "memory/feedback/pol_pplan_not_a_mirror.md", "memory/reference/calculate_installments_endpoint.md", "memory/reference/premium_floor_and_installment_ceiling_surfaces.md", "memory/reference/insis_first_month_due_premium_reversal.md", "knowledge/insis_premium_reversal_first_month.md", "sql/ref/preminst_prod.sql" ], "evidence": "30 tickets asked / 75 analyses used; 32 → 12 h — the largest proportional saving in the catalogue; 4 holders.", "evidence_row": 10 }, { "id": "data.tool.commission-compare", "kind": "tool", "label": "Commission and self-retention comparator", "one_line": "Compare the participant commission rows of both administrations against the effective rule set and emit the correction.", "owner": "support.investigator.pricing", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_agent_commission_prefill.md", "memory/reference/cfg_agent_gen_commission_rules_gap.md", "memory/reference/commission_hold_transfer_comparator_gap.md", "memory/reference/renewal_same_account_agent_drift_report.md", "memory/reference/insis_broker_linked_discount_config.md", "memory/reference/insis_broker_on_direct_ipal_renewal.md" ], "evidence": "16 tickets asked / 45 analyses used; 4 holders; one memory file names the missing comparator explicitly.", "evidence_row": 17 }, { "id": "data.tool.fx-recompute", "kind": "tool", "label": "Currency and rate recomputation", "one_line": "Recompute a policy's amounts against the rate that applied on its rate date and report every row that does not reconcile.", "owner": "support.investigator.pricing", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "knowledge/currency_and_rates.md", "memory/reference/bgn2eur_policy_conversion.md", "memory/reference/bgn_eur_conversion.md", "memory/reference/insis_currency_rate_lookup.md", "memory/reference/cargo_1102_historical_fx_rate_override.md", "memory/reference/cargo_products_foreign_currency_native.md", "memory/reference/myr_euro_conversion_repair_miss.md", "memory/reference/sPPlan_currency_unconditional_insert.md" ], "evidence": "9 tickets asked / 19 analyses used; 8 seeds. The currency changeover is separately an assessment-and-referral operation.", "evidence_row": 28 }, { "id": "data.tool.cargo-statement-correct", "kind": "tool", "label": "Cargo statement and certificate corrector", "one_line": "Apply a corrected value or text to a monthly cargo statement across both administrations and re-total it.", "owner": "support.investigator.transfers", "domain": "data-pricing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/cargo_products.md", "memory/reference/cargo_policy_update.md", "memory/reference/cargo_recalc_alternative_to_unk0103.md", "memory/reference/cargo_1103_expired_parent_with_renewal_discriminator.md", "memory/feedback/customer_text_apply_verbatim.md", "knowledge/insis_ul_cargo_object_edit.md", "memory/reference/cargo_iv_1100_1102_maxv_10m.md" ], "evidence": "1103 statement correction 13 / 21 and cargo text correction 12 / 5 — the text shape is thin at 2 holders.", "evidence_row": 21 }, { "id": "data.skill.print-diagnosis", "kind": "skill", "label": "Diagnose document production", "one_line": "Separate wrong source data from wrong print configuration from a failed invocation, and name which of the three it is.", "owner": "support.investigator.printing", "domain": "data-printing", "status": "existing", "existing_ref": "data.skill.print-diagnosis", "seeds": [ "memory/reference/document_generation.md", "memory/reference/cfg_print_docs_template_configuration.md", "memory/feedback/dont_fallback_on_template_error.md", "memory/reference/insis_print_configuration.md" ], "evidence": "50 tickets asked / 87 analyses used; 124 → 96 h; 4 holders.", "evidence_row": 3 }, { "id": "data.skill.document-production", "kind": "skill", "label": "Produce or retrieve a policy document", "one_line": "Resolve the template and every parameter from the database, evaluate the print gates and produce the document.", "owner": "support.investigator.printing", "domain": "data-printing", "status": "existing", "existing_ref": "data.skill.document-production", "seeds": [ ".claude/skills/print-policy/skill.md", "knowledge/policy_print_recipe.md", "memory/reference/bipublisher_direct_runreport_probe.md", "memory/reference/sr_docs_chain.md" ], "evidence": "Same shape as the diagnosis row: 50 / 87.", "evidence_row": 3 }, { "id": "helper.print-probe", "kind": "tool", "label": "Invocation, registration, data-readiness and artefact-age probe", "one_line": "Answer 'did it run at all, is it registered, is the data ready, how old is the artefact' for one policy and document type.", "owner": "support.investigator.printing", "domain": "data-printing", "status": "existing", "existing_ref": "helper.print-probe", "seeds": [ "memory/reference/document_generation.md", "memory/reference/cfg_print_docs_template_configuration.md", "tools/aisa-blob/README.md", "knowledge/policy_print_recipe.md", "memory/reference/pr_basic_product_documents_lov.md", "memory/reference/gen_risk_documents_validity_change_enddate_gap.md" ], "evidence": "Serves 6 operations in the inventory; 'invocation before configuration' is one of the 13 inherited methodology rows.", "evidence_row": 3 }, { "id": "data.tool.bipublisher-runreport", "kind": "tool", "label": "Direct report-server call", "one_line": "Call the report server directly with resolved parameters and return the document, bypassing the application entirely.", "owner": "support.investigator.printing", "domain": "data-printing", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/bipublisher_direct_runreport_probe.md", "knowledge/policy_print_recipe.md", ".claude/skills/print-policy/skill.md", "memory/reference/serdica_print_service_layout_divergence.md" ], "evidence": "An implementation already exists outside this repository as a local desktop tool with self-test verbs; the recipe here is complete enough to reimplement it.", "evidence_row": 3 }, { "id": "common.tool.aisa-blob", "kind": "tool", "label": "Stored-document extractor", "one_line": "Pull a stored document or large object out of the database by policy, document or blob id and write it to a file.", "owner": "support.investigator.printing", "domain": "data-printing", "status": "existing", "existing_ref": "common.tool.aisa-blob", "seeds": [ "tools/aisa-blob/README.md", "memory/reference/ipal_blob_extraction_via_mcp.md", "memory/reference/fileserver_identifiers_and_store.md", "memory/reference/bulk_document_extraction_windows.md" ], "evidence": "Replaces a batch workaround the memory records as roughly eighty times slower.", "evidence_row": null }, { "id": "data.article.print-surfaces", "kind": "article", "label": "The print surfaces and where they diverge", "one_line": "Which service renders which document, and where the layouts and field sources of the three print paths disagree.", "owner": "support/investigators/printing", "domain": "data-printing", "status": "article", "existing_ref": null, "seeds": [ "memory/reference/serdica_print_service_layout_divergence.md", "memory/reference/serdica_pdfgeneration_docx_templates.md", "memory/reference/bulstrad_online_print_templates.md", "memory/reference/multitrip_print_reads_latin_byclient_address_phone.md", "memory/reference/insis_property_pdf_inverted_mapping.md", "memory/reference/mtpl_doc_attach_flow.md" ], "evidence": "6 memory files describe divergent print paths for the same nominal document.", "evidence_row": null }, { "id": "data.skill.access-diagnosis", "kind": "skill", "label": "Diagnose account and access layers", "one_line": "Walk identity, account, office and role for one person and name the layer that blocks them.", "owner": "support.investigator.access", "domain": "data-access", "status": "existing", "existing_ref": "data.skill.access-diagnosis", "seeds": [ "memory/reference/bulstrad_employee_account.md", "memory/feedback/ipal_user_layer_separation.md", "memory/reference/identity_server_ldap_authentication.md", "memory/feedback/no_forgotten_password.md", "memory/feedback/access_login_pair_db_with_kibana.md" ], "evidence": "38 tickets asked / 62 analyses used; 72 → 42 h; access/account is 8 % of tickets with the highest DB-write rate at 86 %.", "evidence_row": 6 }, { "id": "helper.access-inventory", "kind": "tool", "label": "Identity, account, office and role inventory", "one_line": "Return the whole access tree for one identity and expand the effective audience of a role or route.", "owner": "support.investigator.access", "domain": "data-access", "status": "existing", "existing_ref": "helper.access-inventory", "seeds": [ "memory/reference/identity_server_ldap_authentication.md", "memory/reference/ipal_routes_product_access.md", "memory/feedback/ipal_user_layer_separation.md", ".claude/skills/add-agent/skill.md", "memory/feedback/account_tree.md", "knowledge/roles_and_authorization.md" ], "evidence": "Serves 4 operations; the account and the agent/office shapes together are 73 tickets asked.", "evidence_row": 6 }, { "id": "data.skill.agent-office", "kind": "skill", "label": "Provision an agent, broker or office", "one_line": "Compare the counterpart administration's office and people against the estate's branches, accounts, roles and numbering counters, then fill the gaps.", "owner": "support.investigator.access", "domain": "data-access", "status": "existing", "existing_ref": "data.skill.agent-office", "seeds": [ ".claude/skills/add-agent/skill.md", "memory/reference/ipal_new_office_provisioning_checklist.md", "memory/reference/exclusive_agents.md", "memory/reference/broker_per_office_agent_pattern.md", "memory/reference/bulstrad_office_individual_agent_activation.md", "memory/reference/seq_polno_office_script_superset.md", "memory/reference/ipal_offices_table.md", "sql/proposed/ip_office_branch_user_provisioning_17614_17313_17254.sql", "sql/proposed/provision_ip_18310_kalisto_2016.sql", "memory/project/ip_office_sequence_grants_blocked_backlog.md" ], "evidence": "35 tickets asked / 38 analyses used; 70 → 46 h; 7 one-off provisioning scripts in the checkout repeat one pattern. No catalogue entry, although its two siblings add-bso and add-car have one.", "evidence_row": 8 }, { "id": "data.tool.role-grant-diff", "kind": "tool", "label": "Bulk role grant and revoke difference", "one_line": "Expand a target audience rule to the accounts it selects, diff it against the current grants and emit the idempotent delta with counts.", "owner": "support.investigator.access", "domain": "data-access", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_routes_product_access.md", "memory/reference/ipal_role_usage_inventory.md", "memory/feedback/ur_user_role.md", "memory/reference/ur_12_installments_role.md", "memory/reference/accs_policies_visibility_model.md", "sql/proposed/sd1353_makso_prod_ur_makso_grants.sql", "sql/proposed/sd1954_grant_ur_commercial_property_to_empl.sql", "sql/proposed/sd1353_makso_archive_broad_grants_and_restrict.sql" ], "evidence": "6 tickets asked / 47 analyses used; the one-off scripts in the checkout grant roles to roughly 790 accounts by criteria.", "evidence_row": 30 }, { "id": "data.tool.person-match", "kind": "tool", "label": "Person and company matcher across both administrations", "one_line": "Match one national or company identifier, or a name, to its records in both administrations and flag the duplicates.", "owner": "support.investigator.access", "domain": "data-access", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/ipal_insis_person_matching.md", "memory/reference/ipal_duplicate_customer_merge.md", "memory/reference/insis_p_people_auto3p_duplicate.md", "memory/reference/ipal_customer_sync_flow.md", "memory/reference/c_cust_type_and_addresses_pick.md", "memory/reference/ipal_egn_birthdate_timezone_shift.md", "memory/reference/insis_p_address_language_variants.md", "knowledge/persons.md" ], "evidence": "8 memory/knowledge files; duplicate-customer resolution is its own operation in the inventory.", "evidence_row": null }, { "id": "data.tool.contact-address-dedup", "kind": "tool", "label": "Duplicate contact and address cleanup", "one_line": "Report or merge the duplicate contact and address rows one customer accumulates from delete-and-reinsert churn.", "owner": "support.investigator.access", "domain": "data-access", "status": "candidate", "existing_ref": null, "seeds": [ "sql/maintenance/clean_duplicate_contacts_addresses.sql", "memory/reference/ipal_customer_sync_flow.md", "memory/reference/ipal_insis_id_mapping_tables.md", "memory/reference/c_cust_type_and_addresses_pick.md" ], "evidence": "A parameterised two-phase script already exists in the checkout: statistics first, cleanup second, scopable to one customer.", "evidence_row": null }, { "id": "data.skill.person-policies", "kind": "skill", "label": "List a person's or company's policies", "one_line": "Sweep the four surfaces a person can appear on and return every policy with the cross-system findings.", "owner": "support.resolution", "domain": "data-access", "status": "existing", "existing_ref": "data.skill.person-policies", "seeds": [ ".claude/skills/person-policies/skill.md", "sql/templates/person_policies_json.sql", "memory/feedback/person_policies_fast_path.md", "memory/reference/person_policies_json_template.md" ], "evidence": "5 tickets asked; thin — 1 holder with two or more tickets.", "evidence_row": 33 }, { "id": "data.tool.person-policies-sql", "kind": "tool", "label": "Four-surface policy sweep", "one_line": "One parameterised query returning every policy for one identifier as a single document, with the mismatch findings inline.", "owner": "support.resolution", "domain": "data-access", "status": "existing", "existing_ref": "data.tool.person-policies-sql", "seeds": [ "sql/templates/person_policies_json.sql", "memory/reference/person_policies_json_template.md", "memory/feedback/person_policies_fast_path.md" ], "evidence": "Already an executable template; the shape it serves is thin at 1 holder.", "evidence_row": 33 }, { "id": "data.tool.report-extract", "kind": "tool", "label": "Parameterised customer extract", "one_line": "Run a parameterised select for a customer request and hand back a spreadsheet or delimited file.", "owner": "support.resolution", "domain": "data-access", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/plsql_xlsx_generation_options.md", "sql/templates/person_policies_json.sql", "memory/feedback/fast_set_based_sql_diagnostics.md", "memory/feedback/bulk_loop_output_and_parsing_discipline.md" ], "evidence": "5 tickets asked; thin — 1 holder. The export format question is already resolved in memory.", "evidence_row": 32 }, { "id": "data.skill.health-diagnosis", "kind": "skill", "label": "Diagnose health lifecycle and totals", "one_line": "Follow a health case from census through claim, evaluation and payment and name the step that produced the wrong total.", "owner": "support.investigator.health", "domain": "data-health", "status": "existing", "existing_ref": "data.skill.health-diagnosis", "seeds": [ "bst_health/README.md", "knowledge/health_product_8000_anatomy.md", "memory/reference/health_claim_pipeline_step14_overwrites_step13.md", "memory/reference/health_claims_coinsurance_double_division.md", "memory/reference/health_claim_apply_drug_limits_no_drugs_defect.md" ], "evidence": "No independent ticket count was measured; the health branch is feature-gated to one environment.", "evidence_row": null }, { "id": "data.skill.health-report-review", "kind": "skill", "label": "Review a health provider report", "one_line": "Read a provider report against the check catalogue: the file, its lines, the card holder, the claim behind it and the decision taken.", "owner": "support.investigator.health", "domain": "data-health", "status": "existing", "existing_ref": "data.skill.health-report-review", "seeds": [ ".claude/skills/review-ao-report/skill.md", "memory/reference/hlt_report_line_correction_paths.md", "memory/reference/hlt_ao_report_discount_vs_balance.md", "memory/reference/dzo_group_claims_process.md" ], "evidence": "Qualitative: an established procedure with a per-round test set, no measured recurrence.", "evidence_row": null }, { "id": "helper.health-line-compare", "kind": "tool", "label": "Report line to claim to evaluation to payment comparator", "one_line": "Follow one report line through to its payment and show where the amounts stop agreeing.", "owner": "support.investigator.health", "domain": "data-health", "status": "existing", "existing_ref": "helper.health-line-compare", "seeds": [ ".claude/skills/review-ao-report/skill.md", "memory/reference/hlt_invoice_aggregate_formulas.md", "memory/reference/health_claim_pipeline_step14_overwrites_step13.md", "memory/reference/hlt_imp_data_columns_dont_mean_their_names.md" ], "evidence": "Serves 5 health operations in the inventory.", "evidence_row": null }, { "id": "data.tool.hlt-invoice-recompute", "kind": "tool", "label": "Health invoice aggregate recompute", "one_line": "Recompute the invoice aggregates from their stated formulas and compare them with what is stored, day-boundary drift included.", "owner": "support.investigator.health", "domain": "data-health", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/hlt_invoice_aggregate_formulas.md", "memory/reference/hlt_invoice_period_auto_manual.md", "memory/reference/hlt_invoice_date_utc_day_drift.md", "memory/reference/hlt_ao_report_discount_vs_balance.md" ], "evidence": "4 memory files, one of which writes the aggregate formulas out in full.", "evidence_row": null }, { "id": "data.tool.hlt-import-line-classify", "kind": "tool", "label": "Health import line classifier", "one_line": "Classify every line of an imported provider file as accepted, declined or errored and name the rule that decided it.", "owner": "support.investigator.health", "domain": "data-health", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/hlt_import_declined_vs_error_line_states.md", "memory/reference/hlt_imp_data_columns_dont_mean_their_names.md", "memory/reference/hlt_import_cover_binding_and_insis_10char_rule.md", "memory/reference/hlt_import_person_name_check.md", "memory/reference/hlt_imp_tables_indexes_and_grants.md", "memory/reference/hlt_ao_import_abacus_call_amplification.md" ], "evidence": "6 memory files; one of them is the cause of the repeated-rating-work operation in the inventory.", "evidence_row": null }, { "id": "data.tool.hlt-payment-packet", "kind": "tool", "label": "Health payment confirmation and undo packet", "one_line": "Build the payment confirmation for a report, and its undo, including the mass-payment file the bank expects.", "owner": "support.investigator.health", "domain": "data-health", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/hlt_ao_payment_confirm_undo.md", "memory/reference/hlt_ao_import_batch_layer_and_payment_gaps.md", "memory/reference/eurobank_mass_payment_file_format.md", "memory/project/hlt_ao_export_import_directories.md" ], "evidence": "4 memory files; the file format is a fixed specification and the undo path is documented, which is the teardown a write packet needs.", "evidence_row": null }, { "id": "data.article.health-8000", "kind": "article", "label": "The health product and its pipeline", "one_line": "What product 8000 is: its packages and covers, the census, the claim pipeline, the provider report and the payout.", "owner": "support/investigators/health", "domain": "data-health", "status": "article", "existing_ref": null, "seeds": [ "bst_health/README.md", "knowledge/health_product_8000_anatomy.md", "memory/reference/health_product_8000_cover_modes.md", "memory/reference/dzo_group_claims_process.md", "memory/reference/bst_health_two_company_model_acta.md", "memory/reference/census_member_code_two_provisioning_paths.md", "memory/project/health_products_qa_development.md" ], "evidence": "A self-contained knowledge base of its own plus 20-odd health memory files; five overlapping editions of one domain guide need collapsing on migration.", "evidence_row": null }, { "id": "data.skill.add-bso", "kind": "skill", "label": "Blank range and lifecycle check", "one_line": "Validate a blank number range against its format and the existing stock, add only the free numbers and verify the count.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "existing", "existing_ref": "data.skill.add-bso", "seeds": [ ".claude/skills/add-bso/skill.md", "memory/reference/bso_format.md", "memory/reference/bso_post_issue_maintenance_sequences.md", "memory/reference/insis_blank_registry_rejection.md" ], "evidence": "5 tickets asked / 32 analyses used; 4 holders — not a thin shape, which corrects an earlier reading.", "evidence_row": 31 }, { "id": "data.tool.bso-range-validate", "kind": "tool", "label": "Blank range validator", "one_line": "Check a range for format, script confusion between the two alphabets, overlap and prior use, and emit only the missing numbers.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/bso_format.md", "memory/reference/assistance_card_numbers.md", "memory/reference/bulstrad_online_sales_bso_pool.md", "memory/reference/cfg_blank_numbers_office_extension_proposal.md", "memory/reference/insis_blank_registry_rejection.md", ".claude/skills/add-bso/skill.md" ], "evidence": "The mechanical half of a 5 / 32 shape; the validation rules are fully written out in memory.", "evidence_row": 31 }, { "id": "data.skill.add-car", "kind": "skill", "label": "Vehicle nomenclature alignment", "one_line": "Look a make and model up in all three nomenclatures, decide which of them must change and emit the insert set.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "existing", "existing_ref": "data.skill.add-car", "seeds": [ ".claude/skills/add-car/skill.md", "memory/reference/eisoukr_brand_nomenclature.md", "memory/reference/car_type_mapping.md", "memory/reference/fuel_type_mapping.md" ], "evidence": "2 tickets asked / 2 analyses used; thin — 1 holder.", "evidence_row": 34 }, { "id": "data.tool.car-compare-sql", "kind": "tool", "label": "Vehicle comparator", "one_line": "Field-by-field comparison of a vehicle's record in both administrations by chassis or registration number.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "existing", "existing_ref": "data.tool.car-compare-sql", "seeds": [ "sql/ref/srd_integr_compare_ipal_insis_car.sql", "sql/proposed/compare_ipal_insis_car_rpad_fix.sql", "memory/feedback/insis_ipal_car_sync_invariant.md", "memory/reference/oracle_rpad_truncates_silently.md" ], "evidence": "A deployed procedure today; a patched version fixing its display truncation is in the checkout awaiting promotion.", "evidence_row": 16 }, { "id": "data.tool.vin-chain-update", "kind": "tool", "label": "Chain-aware vehicle correction packet", "one_line": "Emit the correction for a chassis, plate or vehicle type across the object version chain and both administrations, with the comparison as its proof.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/vin_update_chain_aware.md", "memory/reference/vehicle_change_vin_case_split.md", "memory/reference/vehicle_subtype_change_support.md", "memory/reference/dkn_plate_transfer_pattern.md", "memory/reference/engine_no_null_vs_star_marker.md", "memory/feedback/insis_ipal_car_sync_invariant.md", "sql/proposed/SD-1950_reconcile_vin_objects_ipal.sql" ], "evidence": "29 tickets asked once the mail stream is counted / 16 analyses used; moved into build group 1 by D59.", "evidence_row": 16 }, { "id": "data.tool.eisoukr-status-decode", "kind": "tool", "label": "Regulator status and duplicate-cover decoder", "one_line": "Turn a regulator response code and the active-cover check into one named outcome with the blocking policy identified.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/eisoukr_response_codes.md", "memory/reference/eisoukr_mtpl_status_service.md", "memory/reference/duplicate_policy_check_4710.md", "memory/reference/mtpl_duplicate_10_off_open_question.md", "memory/reference/test_eisoukr_simulator.md", "memory/reference/kat_eisoukr_authoritative_car_parameters.md", "sql/ref/srd_integr_tr_lib_motor_prod.sql" ], "evidence": "17 tickets asked / 81 analyses used; 39 → 30 h; 3 holders; the response-code catalogue is already a table.", "evidence_row": 14 }, { "id": "data.tool.vehicle-nomenclature-lookup", "kind": "tool", "label": "Three-nomenclature vehicle lookup", "one_line": "Look one make and model up in all three catalogues at once and report which of them is missing it.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/add-car/skill.md", "memory/reference/eisoukr_brand_nomenclature.md", "memory/reference/kat_getkatdata_soap_pull.md", "memory/reference/car_type_mapping.md", "memory/reference/iis_objects_insis_to_ipal_car_import.md" ], "evidence": "The mechanical core of the thinnest shape in the catalogue: 2 / 2, 1 holder, 100 % lead share.", "evidence_row": 34 }, { "id": "data.tool.blank-after-issue", "kind": "tool", "label": "Issued-blank replacement and invalidation", "one_line": "Replace, cancel or invalidate an already-issued green card or sticker and keep the registry and the regulator in step.", "owner": "support.investigator.master_data", "domain": "data-master", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/bso_post_issue_maintenance_sequences.md", "memory/reference/gc_period_insis_gf_desync.md", "memory/reference/insis_blank_registry_rejection.md", "memory/reference/assistance_card_numbers.md" ], "evidence": "4 memory files; a distinct operation from adding a range, and the only blank operation that touches the regulator.", "evidence_row": null }, { "id": "data.tool.db-health-probe", "kind": "tool", "label": "Database estate health probe", "one_line": "One read-only sweep of tablespace, archiver, blocking sessions, database-link sessions, library-cache pins and scheduler jobs, returning the named condition.", "owner": "support.investigator.operations", "domain": "data-operations", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/insis_tablespace_exhaustion_transfer_halt.md", "memory/reference/insis_archiver_ora00257_policy_no_halt.md", "memory/reference/insis_aud_buffer_busy_outage.md", "memory/reference/insis_library_cache_pin_deadlock.md", "memory/reference/ora_02049_self_block_via_insis_forms.md", "memory/reference/orphaned_dblink_session_transfer_backlog.md", "memory/reference/insis_session_caps_serdica_ablera.md", "knowledge/db_scheduler_jobs.md", "memory/reference/insis_dblink_performance_regression.md" ], "evidence": "9 memory/knowledge files behind 5 operations that share one profile today and have no helper of their own; each signature presents as a data error.", "evidence_row": null }, { "id": "data.tool.idcontrol-headroom", "kind": "tool", "label": "Identifier allocator headroom report", "one_line": "Report every identifier allocator approaching its length overflow, with the percentage consumed.", "owner": "support.investigator.operations", "domain": "data-operations", "status": "candidate", "existing_ref": null, "seeds": [ "sql/diag/idcontrol_headroom_report.sql", "memory/reference/idcontrol_allocate_id_length_overflow.md", "sql/ref/idcontrol.sql", "memory/reference/insis_sequences_per_table.md" ], "evidence": "A runnable whole-estate report exists in the checkout today; it is preventive rather than reactive, so it has no ticket count.", "evidence_row": null }, { "id": "data.tool.transfer-backlog-partition", "kind": "tool", "label": "Transfer backlog partitioner", "one_line": "Split a pending-transfer backlog into steady state, retryable and genuinely stuck, and give each class its action.", "owner": "support.investigator.operations", "domain": "data-operations", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/int_trnsf_pend_bulk_triage.md", "memory/feedback/int_trnsf_pend_steady_state_not_outage.md", "memory/reference/orphaned_dblink_session_transfer_backlog.md", "memory/project/db_transfer_timeouts.md", "memory/feedback/kgas_pileup_not_proof_of_degradation.md" ], "evidence": "5 memory files; two of them exist precisely because a backlog was misread as an outage.", "evidence_row": null }, { "id": "data.tool.mq-probe", "kind": "tool", "label": "Message-bus queue and latency probe", "one_line": "Read queue depth, consumers and per-request latency for the message bus and identify a killed or unbounded channel.", "owner": "support.investigator.operations", "domain": "data-operations", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/rabbitmq_qa_management_access.md", "memory/reference/rabbitmq_per_rid_latency_trace.md", "memory/reference/serdica_rabbit_manual_command_publish.md", "memory/reference/pas_notes_unbounded_reply_amqp_channel_kill.md", "knowledge/microservice_topology.md" ], "evidence": "5 memory/knowledge files; the bus sits between every microservice pair in the topology.", "evidence_row": null }, { "id": "data.article.outage-signatures", "kind": "article", "label": "Infrastructure signatures that look like data errors", "one_line": "The catalogue of platform conditions whose user-visible symptom is a wrong business message.", "owner": "support/investigators/operations", "domain": "data-operations", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/kgas_pileup_not_proof_of_degradation.md", "memory/reference/anl_risk_model_cold_load_rpc_timeout.md", "memory/project/abacus_morning_rush_60s_timeouts.md", "memory/reference/car_price_ws_getpricebyvin_timeout.md", "memory/reference/kuha_polica_gfirm_timeout.md", "memory/reference/apigateway_endpoints_timeout_architecture.md", "memory/reference/insis_p_agents_replica_sync.md", "memory/reference/core_camunda_in_process_bottleneck.md" ], "evidence": "8 memory files; 'a data-shaped error is not proof of a data cause' recurs across all five recovery operations.", "evidence_row": null }, { "id": "configuration.skill.normalization", "kind": "skill", "label": "Understand and reconcile product sources", "one_line": "Turn a specification into a validated product model, with the baselines and the blocking questions named.", "owner": "configuration.normalizer", "domain": "configuration", "status": "existing", "existing_ref": "configuration.skill.normalization", "seeds": [ "knowledge/product_config_new_product.md", "knowledge/product_configuration.md", ".claude/skills/configure-product/skill.md", "memory/feedback/find_the_requirement_before_proposing_change.md" ], "evidence": "Configuration changes close in a median 9 days against 1 for a correction.", "evidence_row": null }, { "id": "configuration.tool.product-validator", "kind": "tool", "label": "Product-model consistency checker", "one_line": "Check an authored product model against its schema, recount its members and resolve its cross-references.", "owner": "configuration.normalizer", "domain": "configuration", "status": "existing", "existing_ref": "configuration.tool.product-validator", "seeds": [ "aisa-next/10 Architecture/20 Configuration/contracts/validate_product.py", "knowledge/product_config_new_product.md" ], "evidence": "Already an executable in the wiki's own contracts folder; it cannot confirm the source's meaning, only the model's self-consistency.", "evidence_row": null }, { "id": "configuration.skill.product-build", "kind": "skill", "label": "Build a product from its specification", "one_line": "Propose the configuration parts a specification implies, agree the scope, then emit the ordered set, the acceptance script and the rollback.", "owner": "configuration.root", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/configure-product/skill.md", "knowledge/product_config_new_product.md", "knowledge/product_configuration.md", "memory/reference/config_key_uniqueness_map.md", "memory/reference/dev_product_landscape.md" ], "evidence": "The v1 twelve-part build is the single largest skill in the workspace; v2 splits it across the five configuration stages, so this row records the seed rather than a new capability.", "evidence_row": null }, { "id": "helper.config-diff", "kind": "tool", "label": "Target-aware configuration diff", "one_line": "Diff schema, rows, effective versions and consuming code for one configuration area between two targets.", "owner": "configuration.ipal.builder", "domain": "configuration", "status": "existing", "existing_ref": "helper.config-diff", "seeds": [ "memory/feedback/note_incidental_config_diffs_dont_normalize.md", "memory/reference/config_key_uniqueness_map.md", "memory/reference/null_semantics_per_config_table.md", "memory/reference/config_surface_liveness_test.md", "knowledge/product_configuration.md" ], "evidence": "Serves 7 configuration operations in the inventory.", "evidence_row": null }, { "id": "helper.numbering-preflight", "kind": "tool", "label": "Non-consuming numbering preflight", "one_line": "Check rows, sequence, grant, synonym, principal and collisions for a numbering configuration without consuming a number.", "owner": "configuration.serdica.builder", "domain": "configuration", "status": "existing", "existing_ref": "helper.numbering-preflight", "seeds": [ "memory/reference/policy_no_generation.md", "memory/reference/manual_policy_no_preconsumes_sequence_collision.md", "memory/reference/seq_polno_office_script_superset.md", "memory/reference/policy_no_composition.md", "memory/reference/policy_no_office_frozen_at_first_transfer.md", "memory/reference/cargo_child_number_shared_series_collision.md" ], "evidence": "6 tickets asked / 14 analyses used; serves the numbering and the number-collision operations.", "evidence_row": 29 }, { "id": "configuration.tool.validation-cell-toggle", "kind": "tool", "label": "Validation threshold relax and revert", "one_line": "Capture the baseline of a validation cell, relax it for one bounded operation and revert it to the captured baseline.", "owner": "configuration.ipal.builder", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/validation_threshold_temporary_relaxation.md", "memory/reference/field_validation_frameworks.md", "memory/feedback/validation_config_freshness.md", "memory/reference/cfg_manual_premium_control.md", "memory/feedback/check_abacus_layer_for_validations.md", "memory/reference/abacus_validation_pipeline.md", "memory/feedback/never_promote_config_ahead_of_evaluator.md" ], "evidence": "18 tickets asked / 41 analyses used; 4 holders. It is the clearest write-with-teardown in the corpus.", "evidence_row": 13 }, { "id": "configuration.tool.label-locate", "kind": "tool", "label": "Label, message and translation locator", "one_line": "Find which of the three text surfaces owns a displayed string and emit the per-language update for it.", "owner": "configuration.serdica.builder", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_descr_link_chain.md", "memory/reference/ui_i18n_vs_sr_messages.md", "memory/reference/sr_messages_before_lt_doc_codes.md", "memory/reference/pricing_factor_dependent_display_label.md", "memory/reference/insys_messages_reins_79.md" ], "evidence": "12 tickets asked / 27 analyses used; 4 holders; shared with the configuration label sweep.", "evidence_row": 25 }, { "id": "configuration.tool.factor-value-add", "kind": "tool", "label": "Pricing factor and dependent-choice add", "one_line": "Add a factor value, and its dependency key where the factor is dependent, to every product that needs it, from the counterpart's source row.", "owner": "configuration.ipal.builder", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_pricing_factor_framework.md", "memory/reference/abc_cfg_pricing_factors_mapping.md", "memory/reference/graphql_prpricingfactorvalues.md", "memory/reference/2215_home_add_dependent_lov_ui_gap.md", "memory/reference/pricing_factor_dependent_display_label.md" ], "evidence": "11 tickets asked / 42 analyses used; 4 holders; the flat-versus-dependent trap has its own memory file.", "evidence_row": 26 }, { "id": "configuration.tool.tariff-table-check", "kind": "tool", "label": "Official tariff table check", "one_line": "Check a computed premium against the transcribed official tariff matrix for the product, band by band.", "owner": "configuration.source_analyst", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ "knowledge/motor_tariff_2026.md", "memory/reference/4710_abacus_factor_defaults.md", "memory/reference/4704_casco_covers_factors.md", "memory/reference/4710_electric_vehicle_tariff_band_gap.md", "memory/reference/casco_no_claims_discount_years_based.md" ], "evidence": "The 2026 motor matrices are already transcribed as fixed tables; the source analyst is a profile denied a database, so a table it can read locally is exactly its instrument.", "evidence_row": null }, { "id": "configuration.tool.print-doc-register", "kind": "tool", "label": "Registered document-set change", "one_line": "Add, remove or re-point a document registration for a product and prove the registration is what the print gate reads.", "owner": "configuration.serdica.builder", "domain": "configuration", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/cfg_print_docs_template_configuration.md", "memory/reference/pr_basic_product_documents_lov.md", "memory/reference/insis_print_configuration.md", "memory/reference/sr_docs_chain.md" ], "evidence": "4 memory files; the diagnostic side is covered by the print probe, the registration side is not.", "evidence_row": null }, { "id": "configuration.article.config-surfaces", "kind": "article", "label": "Where product behaviour is configured", "one_line": "Every configuration surface across the three systems that decides how a product behaves, and which key is unique where.", "owner": "configuration", "domain": "configuration", "status": "article", "existing_ref": null, "seeds": [ "knowledge/product_configuration.md", "knowledge/product_config_new_product.md", "memory/reference/config_key_uniqueness_map.md", "memory/reference/null_semantics_per_config_table.md", "memory/reference/config_surface_liveness_test.md", "knowledge/abacus_schema.md", "knowledge/abacus_engine_vs_abc_access_schema.md" ], "evidence": "7 knowledge/memory files; two of them exist only to stop the pricing engine and its resident schema being confused.", "evidence_row": null }, { "id": "configuration.article.product-catalogue", "kind": "article", "label": "The products and what each one is made of", "one_line": "Per-product cover sets, factors, packages and known configuration gaps, product family by product family.", "owner": "configuration", "domain": "configuration", "status": "article", "existing_ref": null, "seeds": [ "memory/reference/4704_casco_covers_factors.md", "memory/reference/cargo_products.md", "memory/reference/product_1126.md", "memory/reference/marsh_3618_packages.md", "memory/reference/lb_offers_and_cover_rules.md", "memory/reference/dev_product_landscape.md", "memory/reference/travel_3618_iv_authority_ladder.md", "memory/reference/vessel_object_model_1126.md" ], "evidence": "Roughly 40 product-specific memory files; 8 are cited here as the shape of the node.", "evidence_row": null }, { "id": "source.skill.plsql-change", "kind": "skill", "label": "Prepare a stored-code change", "one_line": "Snapshot the live body, take its content signature, merge three ways and make the signature a precondition of the apply.", "owner": "source.implementer.db-plsql", "domain": "source", "status": "existing", "existing_ref": "source.skill.plsql-change", "seeds": [ "memory/feedback/db_resident_procedure_fix_workflow.md", "memory/reference/odp_net_plsql_bindbyname.md", "memory/feedback/db_comments_in_english.md", "memory/feedback/db_change_5step_commit_last.md" ], "evidence": "One of the three worked skill samples in the wiki; the workflow is written out in memory.", "evidence_row": null }, { "id": "source.skill.release-audit", "kind": "skill", "label": "Check the serving release", "one_line": "Prove which revision is actually serving on a target, separating deploy noise from a stale build.", "owner": "source.deployer", "domain": "source", "status": "existing", "existing_ref": "source.skill.release-audit", "seeds": [ "memory/reference/test_prod_release_audit.md", "memory/feedback/stack_line_numbers_prove_stale_build.md", "memory/reference/serdica_deploy_restart_noise_patterns.md", "memory/reference/serdica_backend_branch_per_env.md" ], "evidence": "4 memory files; the stale-build signature is a recurring false diagnosis.", "evidence_row": null }, { "id": "source.skill.port", "kind": "skill", "label": "Port only the real delta", "one_line": "Compute the genuine difference between two branch heads and carry only that, checking the base commit first.", "owner": "source.deployer", "domain": "source", "status": "existing", "existing_ref": "source.skill.port", "seeds": [ "memory/feedback/cross_branch_port_check_base_commit.md", "memory/feedback/merge_to_ancestor_branch_is_fast_forward.md", "memory/reference/serdica_branches.md", "memory/feedback/head_moved_under_you_recheck_source.md" ], "evidence": "4 memory files; the branch-per-environment flow makes this a step of every change.", "evidence_row": null }, { "id": "source.tool.plsql-unwrap", "kind": "tool", "label": "Wrapped package source recovery", "one_line": "Return readable source for a stored package whose body is wrapped in the database, from the captured clear copies.", "owner": "common.db-extractor", "domain": "source", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/policyg_unwrapped_source.md", "memory/reference/gen_appl_pkg_unwrapped.md", "sql/ref/policyg.sql", "sql/ref/gen_appl_pkg.sql", "sql/ref/preminst_prod.sql", "sql/ref/gen_install_prod.sql" ], "evidence": "6 seeds; 11 further package bodies are spooled in the health knowledge base's own PL/SQL folder.", "evidence_row": null }, { "id": "source.tool.error-to-throwsite", "kind": "tool", "label": "Error text to throw site", "one_line": "Map a displayed error string to its message id and from there to the exact throw site in stored code or backend source.", "owner": "source.planner", "domain": "source", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/ipal_error_message_to_throw_site.md", "memory/feedback/source_attribution_for_db_call_sites.md", "memory/feedback/disjunctive_error_template_third_source.md", "memory/reference/generic_error_check_object_values_first.md", "knowledge/transfer_pipeline.md", "memory/feedback/db_error_contract.md" ], "evidence": "6 seeds; the transfer pipeline document already carries line-numbered throw sites, which is the lookup table this tool needs.", "evidence_row": null }, { "id": "source.tool.branch-delta", "kind": "tool", "label": "Branch delta and base-commit check", "one_line": "Compute the true delta between two branch heads, verify the base commit and refuse a port whose head moved.", "owner": "source.deployer", "domain": "source", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/cross_branch_port_check_base_commit.md", "memory/feedback/merge_to_ancestor_branch_is_fast_forward.md", "memory/feedback/head_moved_under_you_recheck_source.md", "memory/reference/gitlab_mr_creation_limits.md", "memory/reference/serdica_backend_branch_per_env.md" ], "evidence": "5 memory files; the mechanical half of the existing port skill.", "evidence_row": null }, { "id": "source.article.backend-shape", "kind": "article", "label": "The backend's shape and its traps", "one_line": "The microservice topology, the authoritative compiled model, the mapping profiles and the construction traps.", "owner": "source/serdica-backend", "domain": "source", "status": "article", "existing_ref": null, "seeds": [ "knowledge/microservice_topology.md", "memory/reference/ipal_service_topology.md", "memory/reference/serdica_compiled_ef_model_is_authoritative.md", "memory/reference/bstintegration_automapper_profiles.md", "memory/feedback/no_cross_microservice_references.md", "memory/reference/baseresult_exception_semantics.md", "memory/reference/activator_createinstance_exact_arity.md", "memory/reference/apigateway_endpoints_timeout_architecture.md" ], "evidence": "8 knowledge/memory files; the per-microservice child nodes are auto-created by the summarizer.", "evidence_row": null }, { "id": "source.article.build-and-test", "kind": "article", "label": "Building, testing and running it locally", "one_line": "What it takes to build, seed, run and debug the estate's code on a developer machine, and where the toolchain breaks.", "owner": "source", "domain": "source", "status": "article", "existing_ref": null, "seeds": [ "memory/reference/serdica_scaffold_optimize_pipeline.md", "memory/reference/claims_local_run_pluginbinaries.md", "memory/reference/claims_plugin_host_service_exposure.md", "memory/feedback/serdica_inmemory_ef_test_seeding.md", "memory/feedback/lsp_diagnostics_unreliable_in_serdica_backend.md", "memory/reference/debian11_eol_breaks_riskmodel_docker_build.md", "memory/reference/gitlab_mr_creation_limits.md" ], "evidence": "7 memory files; one operation in the inventory is exactly 'repair a customer build or deployment dependency'.", "evidence_row": null }, { "id": "source.article.db-estate", "kind": "article", "label": "The database estate and its packages", "one_line": "The schemas, tables and stored packages of both administrations, with the comment-versus-reality defects called out.", "owner": "source", "domain": "source", "status": "article", "existing_ref": null, "seeds": [ "knowledge/ipal_schema.md", "knowledge/insis_schema.md", "knowledge/db_overview_and_improvements.md", "knowledge/db_packages.md", "knowledge/abacus_schema.md", "memory/reference/oracle_sql_gotchas.md", "memory/reference/sqlcl_cross_db_sql_quirks.md", "memory/reference/insis_sequences_per_table.md" ], "evidence": "8 knowledge/memory files covering 7 schemas, 210 tables and 95 packages.", "evidence_row": null }, { "id": "commercial.skill.assess-effort", "kind": "skill", "label": "Assess required work and effort", "one_line": "Investigate what a request would take through read-only evidence and record the range, dependencies, exclusions and uncertainty.", "owner": "commercial.assessor", "domain": "commercial", "status": "existing", "existing_ref": "commercial.skill.assess-effort", "seeds": [ "knowledge/support_contract_bg.md", "memory/reference/support_contract_custom_en.md", "memory/feedback/insis_change_request.md" ], "evidence": "22 of 485 tickets are CR candidates; three operations in the inventory route to assessment rather than delivery.", "evidence_row": null }, { "id": "commercial.tool.effort-comparable", "kind": "tool", "label": "Comparable-work lookup for an estimate", "one_line": "Find past cases of the same shape and return their recorded effort and outcome as the comparables behind an estimate.", "owner": "commercial.assessor", "domain": "commercial", "status": "candidate", "existing_ref": null, "seeds": [ "scripts/mine_tickets_classify.py", "scripts/mine_tickets_timing.py", "memory/reference/kb_memory_vs_rediscovery_cost.md", "aisa-next/10 Architecture/40 Support/Operational Method Evidence.md" ], "evidence": "The mining scripts that produced the measured catalogue already exist; 185 reliable timings against 230 human estimates.", "evidence_row": null }, { "id": "commercial.tool.integration-sample-pack", "kind": "tool", "label": "Partner integration sample pack", "one_line": "Emit a partner-ready request and response set for one product from what the product declares, with the pricing factors listed.", "owner": "commercial.assessor", "domain": "commercial", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/register_policy_dto.md", "memory/reference/register_policy_one_shot_from_db.md", "memory/reference/graphql_prpricingfactorvalues.md", ".claude/skills/bulsi/skill.md", "memory/reference/4710_abacus_factor_defaults.md", "memory/reference/bul_si_integration_partner.md" ], "evidence": "6 seeds; one live partner engagement drove all of them and the one-shot builder is already written out.", "evidence_row": null }, { "id": "commercial.article.partner-engagement", "kind": "article", "label": "How a partner integration is run", "one_line": "What a third-party developer needs, what may be shared, and the discipline of answering a partner's question.", "owner": "commercial", "domain": "commercial", "status": "article", "existing_ref": null, "seeds": [ ".claude/skills/bulsi/skill.md", "memory/reference/bul_si_integration_partner.md", "memory/feedback/bulsi_engagement_workflow.md", "memory/feedback/partner_engagement_discipline.md", "memory/feedback/verify_bulsi_mechanics_before_asserting.md", "memory/feedback/no_modify_partner_shared_content.md", "memory/reference/bulsi_3636_premium_rounding_363626001199130.md" ], "evidence": "7 memory files and one v1 skill from a single partner engagement.", "evidence_row": null }, { "id": "common.tool.aisa-jira", "kind": "tool", "label": "Ticket-system client", "one_line": "Fetch, search, comment, attach, tag and transition tickets from the command line, with a double-post guard on internal notes.", "owner": "common.communicator", "domain": "common", "status": "existing", "existing_ref": "common.tool.aisa-jira", "seeds": [ "tools/aisa-jira/README.md", "memory/feedback/jira_via_aisa_jira_cli.md", "memory/feedback/aisa_internal_comment_single_per_ticket.md", "memory/reference/jira_attachment_fetch.md" ], "evidence": "Source and a Release build exist; no build is present at the publish path every skill names.", "evidence_row": null }, { "id": "common.tool.aisa-mail", "kind": "tool", "label": "Mailbox archival client", "one_line": "Sync, slice and extract mail threads from the shared mailbox with a body slicer and inline-answer detection.", "owner": "common.communicator", "domain": "common", "status": "existing", "existing_ref": "common.tool.aisa-mail", "seeds": [ "tools/aisa-mail/README.md", "memory/reference/aisa_mail_cli.md", "memory/feedback/email_analysis_aisa_mail_first.md", "memory/reference/ms_graph_cli_email_fetch_workflow.md" ], "evidence": "Recorded as roughly a 95 % token reduction against the connector for bulk archival.", "evidence_row": null }, { "id": "common.tool.aisa-hdesk", "kind": "tool", "label": "Customer helpdesk client", "one_line": "Authenticate, fetch and sync tickets from the customer's own helpdesk and hand them to the ticket system.", "owner": "support.root", "domain": "common", "status": "existing", "existing_ref": "common.tool.aisa-hdesk", "seeds": [ "tools/aisa-hdesk/README.md", "memory/reference/aisa_hdesk_cli.md", "memory/reference/windows_dpapi_credential_cache.md" ], "evidence": "Source only; no build output exists in the checkout at all.", "evidence_row": null }, { "id": "common.tool.kibana-trace", "kind": "tool", "label": "Log trace by request, policy or user", "one_line": "Query the environment's log stack for one request id, policy or user and return the ordered error chain as evidence.", "owner": "common.db-prober", "domain": "common", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/kibana_log_access.md", "memory/feedback/kibana_guid_to_operator_via_requestid.md", "memory/feedback/kibana_batch_response_printdocs_array.md", "memory/feedback/access_login_pair_db_with_kibana.md", "memory/feedback/db_trace_authoritative_over_net_log.md" ], "evidence": "45 tickets asked / 45 analyses used; 162 → 131 h; the thinnest large shape, 2 holders and an 87 % lead share.", "evidence_row": 4 }, { "id": "common.tool.attachment-fetch", "kind": "tool", "label": "Attachment normaliser", "one_line": "Take an attachment from any arrival channel and return its text, its metadata and its embedded images as case papers.", "owner": "common.communicator", "domain": "common", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/jira_attachment_fetch.md", "memory/reference/outlook_attachment_fetch.md", "memory/reference/local_msg_extraction.md", "memory/reference/jira_video_attachment_frame_extraction.md", "memory/reference/office_docs_read_and_render.md", "memory/reference/bulk_document_extraction_windows.md", "tools/email-extractor/README.md" ], "evidence": "7 seeds across three channels; the legacy extractor is retained precisely because no single normaliser exists.", "evidence_row": null }, { "id": "common.tool.case-corpus-mine", "kind": "tool", "label": "Case-archive miner", "one_line": "Walk the case archive and emit the per-shape counts, effort classes, holders and timings that rank the capability list.", "owner": "common.experience-curator", "domain": "common", "status": "candidate", "existing_ref": null, "seeds": [ "scripts/mine_tickets_extract.py", "scripts/mine_tickets_classify.py", "scripts/mine_tickets_holders.py", "scripts/mine_tickets_timing.py", "scripts/analyze_tickets.py" ], "evidence": "Five working scripts; they produced every measured number the capability catalogues rank on.", "evidence_row": null }, { "id": "common.tool.memory-index-check", "kind": "tool", "label": "Memory index integrity check", "one_line": "Verify that every index entry resolves to a node, every node is indexed and no description contradicts its file.", "owner": "common.experience-curator", "domain": "common", "status": "candidate", "existing_ref": null, "seeds": [ "scripts/audit-knowledge.ps1", "memory/feedback/memory_system_maintenance.md", "memory/feedback/one_description_per_thing.md", "memory/feedback/name_pattern_with_plain_description.md" ], "evidence": "A link-integrity gate already exists for the file-based corpus; 565 memory files and 27 knowledge files depend on it.", "evidence_row": null }, { "id": "common.tool.bg-style-lint", "kind": "tool", "label": "Outbound Bulgarian text lint", "one_line": "Check a drafted reply against the recorded terminology, currency, date and enumeration rules before it is sent.", "owner": "common.communicator", "domain": "common", "status": "candidate", "existing_ref": null, "seeds": [ "memory/feedback/bg_prose_style.md", "memory/feedback/bg_made_up_words.md", "memory/feedback/polica_vs_policy_terminology.md", "memory/feedback/product_4710_terminology.md", "memory/feedback/latin_letters_for_enumerations.md", "memory/feedback/currency_display.md", "memory/feedback/date_format_convention.md", "memory/feedback/bulstrad_life_full_name.md", "memory/feedback/product_names_from_insis.md", "memory/feedback/joro_zhoro_disambiguation.md" ], "evidence": "10 feedback files, each recording a correction the customer or the user made to outbound text.", "evidence_row": null }, { "id": "common.skill.experience", "kind": "skill", "label": "Retrieve and judge experience for this task", "one_line": "Select compatible past cases and knowledge for the task at hand and decide whether they still apply.", "owner": "common.experience-curator", "domain": "common", "status": "existing", "existing_ref": "common.skill.experience", "seeds": [ "memory/reference/kb_memory_vs_rediscovery_cost.md", "memory/feedback/cross_check_and_reanalysis.md", "KNOWN_ISSUES.md", "scripts/triage_baseline.py" ], "evidence": "Cached knowledge measured at 10-60x cheaper and 30-300x faster than rediscovery.", "evidence_row": null }, { "id": "common.skill.retained-learning", "kind": "skill", "label": "Retain and improve a capability", "one_line": "Turn what a case taught into the memory entry, the amended procedure and the executable that stops the next case relearning it.", "owner": "common.experience-curator", "domain": "common", "status": "existing", "existing_ref": "common.skill.retained-learning", "seeds": [ "memory/feedback/propagate_learnings_after_ticket.md", "memory/feedback/auto_update_knowledge.md", "memory/feedback/memory_system_maintenance.md", "memory/feedback/recurring_manual_fix_escalate_to_config.md", "memory/feedback/durable_fix_over_comment.md" ], "evidence": "5 feedback files; the pain point this addresses is stated as the reason for the whole precipitation stage.", "evidence_row": null }, { "id": "common.article.evidence-discipline", "kind": "article", "label": "Evidence discipline", "one_line": "Verify rather than infer: what counts as a finding, what a failed probe means and how a baseline is chosen.", "owner": "support/methodologies", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/verify_dont_guess.md", "memory/feedback/failed_probe_is_not_a_finding.md", "memory/feedback/cross_check_and_reanalysis.md", "memory/feedback/db_investigation_discipline.md", "memory/feedback/baseline_same_state_peer_not_end_state.md", "memory/feedback/ui_panel_zero_not_evidence_of_absence.md", "memory/feedback/dont_infer_clobber_from_snapshot_delta.md", "memory/feedback/dont_infer_staleness_from_format.md", "memory/feedback/no_endpoint_is_not_no_writer.md", "memory/feedback/dont_over_interpret_warning_level_retry_noise.md" ], "evidence": "Roughly 25 feedback files; 6 of the 13 rows of the inherited methodology catalog belong to this family.", "evidence_row": null }, { "id": "common.article.write-discipline", "kind": "article", "label": "Write and remedy discipline", "one_line": "How a change to live data is shown, applied, verified and committed, and when a workaround is preferred to a repair.", "owner": "support/methodologies", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/db_change_5step_commit_last.md", "memory/feedback/strict_read_only_no_rollback.md", "memory/feedback/confirm_before_destructive_actions.md", "memory/feedback/operator_sql_complete_first_pass.md", "memory/feedback/offticket_policy_changes_need_approval.md", "memory/feedback/insis_dml_executor_and_username.md", "memory/feedback/transient_vs_structural_fix.md", "memory/feedback/prefer_return_to_application_for_dual_system_corrections.md", "memory/feedback/prefer_path_a_when_no_regulatory_tie.md", "memory/feedback/decouple_over_delete.md", "memory/feedback/temp_tables.md" ], "evidence": "Roughly 20 feedback files; 220 of 333 analyses end in a database write, which is what this family governs.", "evidence_row": null }, { "id": "common.article.precipitation-discipline", "kind": "article", "label": "Precipitation discipline", "one_line": "When a finding becomes a memory entry, a register row or an escalation, and how the closing walk is reconciled.", "owner": "support/methodologies", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/propagate_learnings_after_ticket.md", "memory/feedback/auto_update_knowledge.md", "memory/feedback/maintain_known_issues_register.md", "memory/feedback/memory_system_maintenance.md", "memory/feedback/one_description_per_thing.md", "memory/reference/kb_memory_vs_rediscovery_cost.md", "memory/feedback/aisa_self_discipline_at_write_time.md" ], "evidence": "7 feedback files; the 20.05.2026 audit measured these very rules failing at author time.", "evidence_row": null }, { "id": "common.article.channel-practice", "kind": "article", "label": "Channel practice", "one_line": "How each arrival channel behaves, what it hides and what must be scanned before a cause is proposed.", "owner": "platform/connectors", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/scan_mailbox_and_microservice_logs_per_subject.md", "memory/feedback/cross_channel_correlation.md", "memory/reference/email_archive_convention.md", "memory/feedback/jira_adf_date_node_invisible_in_list.md", "memory/reference/hdesk_jira_provenance_staleness_check.md", "memory/feedback/non_bulsi_emails_archive_to_emails_dir.md" ], "evidence": "6 memory files; the step-zero cross-channel scan is one of the 13 inherited methodology rows.", "evidence_row": null }, { "id": "platform.article.environment-registry", "kind": "article", "label": "The environments and how they are addressed", "one_line": "Which target is which, the two addresses every database host answers on, and the default environment per kind of work.", "owner": "platform", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/reference/db_host_dual_address.md", "memory/project/sqlcl_connections_state.md", "memory/reference/serdica_backend_branch_per_env.md", "memory/reference/bst_dev_env_and_product_8000_generation.md", "memory/feedback/verify_prod_object_state_on_prod_connection.md", "memory/feedback/work_directly_on_bulstrad_qa.md", "memory/feedback/qa_data_volatile_not_ground_truth.md", ".claude/skills/workspace-initialize/skill.md" ], "evidence": "8 seeds; 'environment defaults' is one of the 13 inherited methodology rows and the dual address has its own file.", "evidence_row": null }, { "id": "platform.article.data-protection", "kind": "article", "label": "Personal data and secrets", "one_line": "What may not enter a stored note, how credentials are held and how a synthetic identifier is made for a test.", "owner": "platform", "domain": "common", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/gdpr_no_customer_pii_in_memory.md", "memory/feedback/secrets_discipline.md", "memory/reference/windows_dpapi_credential_cache.md", "memory/reference/egn_synthetic_generation_rules.md", "memory/reference/ablera_test_accounts.md", "memory/feedback/anthropic_api_call_credential_path.md" ], "evidence": "6 memory files; the synthetic-identifier rules are precise enough to be an executable generator on their own.", "evidence_row": null }, { "id": "support.investigator.operations", "kind": "agent", "label": "Operations and infrastructure investigator", "one_line": "Investigate a symptom whose cause is the platform rather than the data: exhausted storage, blocked sessions, a halted process, a frozen replica.", "owner": "support.investigator.operations", "domain": "data-operations", "status": "candidate", "existing_ref": null, "seeds": [ "memory/reference/insis_tablespace_exhaustion_transfer_halt.md", "memory/reference/orphaned_dblink_session_transfer_backlog.md", "memory/reference/ssh_mcp_cli.md", "memory/reference/insis_session_caps_serdica_ablera.md", "memory/reference/rabbitmq_qa_management_access.md", "memory/feedback/shared_infra_attribution.md", "memory/project/db_transfer_timeouts.md" ], "evidence": "5 operations in the inventory carry the operational-recovery domain and are all assigned to the transfer investigator today; the shell connector they need is not read-only.", "evidence_row": null }, { "id": "workspace.tool.analysis-render", "kind": "tool", "label": "Ticket analysis HTML renderer", "one_line": "Inline the shared stylesheet, script and clause library into a written analysis page and compute its asset depth.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ "scripts/finalize_analysis.ps1", "memory/feedback/ticket_analysis.md", "memory/feedback/html_ticket_style.md", "tools/analysis-assets/aisa_analysis.css", ".claude/skills/_shared/ticket_analysis_template.html" ], "evidence": "The v2 case record replaces the artefact this renders; the renderer serves the file-based workspace only.", "evidence_row": null }, { "id": "workspace.skill.archive-discussion", "kind": "skill", "label": "Archive a discussion thread", "one_line": "Turn a non-incident thread into a self-contained browsable archive with every source file cited and openable.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/archive-discussion/skill.md", "memory/reference/others_archive_convention.md", "memory/reference/email_archive_convention.md", "memory/feedback/discussion_vs_issue_analysis_format.md", "memory/reference/analysis_format_parity_jira_email_others.md" ], "evidence": "In v2 a thread is a case paper, so the archive format itself is not a runtime capability.", "evidence_row": null }, { "id": "workspace.skill.design-docs", "kind": "skill", "label": "Produce a design or specification suite", "one_line": "Run an approval-gated build of a numbered specification suite or a combined design document with a running log.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/design-docs/skill.md", "memory/reference/design_document_suite_convention.md", "memory/reference/readme_html_ceo_doc_trio.md", "memory/reference/mermaid_offline_inline_svg.md" ], "evidence": "Authoring, not operating; the estate's own change documents are Development module outputs.", "evidence_row": null }, { "id": "workspace.tool.token-audit", "kind": "tool", "label": "Assistant configuration token audit", "one_line": "Measure prompt load, tool surface, model switches and the cache split of an assistant configuration and rank the waste.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/token-audit/skill.md", "memory/user/optimization_philosophy.md", "memory/feedback/aisa_disconnect_idle_mcp_sessions.md", "memory/feedback/subagent_session_limit_check_disk_first.md" ], "evidence": "Superseded in v2 by the ledger, which records provider usage as a fact rather than a measured audit.", "evidence_row": null }, { "id": "workspace.tool.wiki-gate", "kind": "tool", "label": "Documentation convention gate", "one_line": "Check the design wiki's links, tags, sources and reachability and regenerate its open-item register.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ "scripts/wiki_check.ps1", "scripts/wiki_challenges.ps1", "memory/feedback/aisa_next_documentation_style.md", "memory/feedback/md_heading_anchors_no_emoji.md" ], "evidence": "Serves this wiki, not the runtime; its memory-tree cousin is listed as a runtime candidate separately.", "evidence_row": null }, { "id": "workspace.tool.hdesk-offline-viewer", "kind": "tool", "label": "Offline helpdesk ticket viewer", "one_line": "Generate a dependency-free local viewer and a dated queue report for the cached helpdesk tickets.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ "hdesk/scripts/gen_cached_views.ps1", "hdesk/scripts/gen_status.ps1", "memory/feedback/hdesk_current_status_html.md", "memory/reference/offline_local_html_viewer.md" ], "evidence": "A cache-browsing convenience; in v2 the helpdesk is a connector and its tickets are cases.", "evidence_row": null }, { "id": "workspace.skill.workspace-initialize", "kind": "skill", "label": "Set the workspace up", "one_line": "Check the runtime prerequisites, install the database client, register the named connections and prebuild the channel clients.", "owner": "workspace", "domain": "workspace", "status": "candidate", "existing_ref": null, "seeds": [ ".claude/skills/workspace-initialize/skill.md", "memory/project/workspace_tooling_state_2026_06.md", "memory/project/sqlcl_connections_state.md", "memory/reference/claude_settings_layout.md" ], "evidence": "Deterministic provisioning of one operator's machine; v2 provisions connectors centrally.", "evidence_row": null }, { "id": "workspace.article.authoring-conventions", "kind": "article", "label": "Authoring conventions", "one_line": "How this workspace's own documents, diagrams, citations and commits are written.", "owner": "workspace", "domain": "workspace", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/aisa_next_documentation_style.md", "memory/reference/mermaid_er_sequence_label_rules.md", "memory/reference/mermaid_class_name_collisions.md", "memory/reference/mermaid_gantt_syntax.md", "memory/feedback/cite_source_document_small_text.md", "memory/feedback/cite_attachments_with_filesystem_link.md", "memory/feedback/commit_messages_no_wrap.md", "memory/feedback/git_commit_only_on_request.md", "memory/feedback/docs_folder_for_on_request_documentation.md" ], "evidence": "Roughly 25 feedback and reference files about writing, diagramming and committing in this repository.", "evidence_row": null }, { "id": "workspace.article.host-platform", "kind": "article", "label": "The host platform's quirks", "one_line": "Shell, encoding, path and tooling behaviour of the machine this workspace runs on.", "owner": "workspace", "domain": "workspace", "status": "article", "existing_ref": null, "seeds": [ "memory/feedback/no_python_for_text_on_windows.md", "memory/feedback/console_utf8_codepage.md", "memory/reference/git_bash_windows_path_conversion.md", "memory/feedback/transient_files_in_tmp.md", "memory/feedback/cli_diagnostics_to_stdout.md", "memory/feedback/tools_dir_matches_binary_name.md", "memory/feedback/script_and_naming_discipline.md", "memory/reference/winforms_layout_traps_and_offscreen_check.md" ], "evidence": "8 feedback and reference files that exist only because the operator machine is a Windows box.", "evidence_row": null }, { "id": "workspace.article.assistant-operation", "kind": "article", "label": "Operating the assistant itself", "one_line": "How this assistant's own configuration, sessions, connectors and identity are handled.", "owner": "workspace", "domain": "workspace", "status": "article", "existing_ref": null, "seeds": [ "memory/user/aisa_agent_identity.md", "memory/user/profile.md", "memory/user/optimization_philosophy.md", "memory/feedback/claude_md_import_syntax.md", "memory/reference/claude_settings_layout.md", "memory/reference/claude_session_url_to_local_transcript.md", "memory/feedback/sqlcl_mcp_run_sql_silent_failure.md", "memory/reference/sqlcl_cli_batch_mining.md" ], "evidence": "The three user files plus 10-odd assistant-operation files; superseded by the platform's own profile and runtime.", "evidence_row": null } ] }