aidp-fusion-autopilotEnd-to-end conductor for the Fusion to AIDP to OAC dashboard journey. Takes one high-level goal ('I want a supplier-spend vs GL-balance dashboard from Fusion') and drives the whole chain: configure, connect OAC MCP, bootstrap, seed, advise dataset, author mart if needed, create OAC dataset, author w
aidp-fusion-bootstrapGuide and run the Oracle AIDP Fusion Autopilot bootstrap phase. Validates bundle/config readiness, checks AIDP/Fusion/IAM prerequisites, runs `aidp-fusion-autopilot bootstrap --check-iam` or `bootstrap --refresh`, surfaces tenant variation choices pinned into the active profile YAML and evidence sna
aidp-fusion-configGenerate aidp.config.yaml from human-friendly AIDP names instead of hand-copied OCIDs. Use when the user is setting up the fusion-autopilot and doesn't want to dig workspaceKey / clusterKey / aiDataPlatformId out of console URLs — they give a workspace name + cluster name (+ region) and this resolve
aidp-fusion-incrementalTurn a natural-language refresh request into a guarded `aidp-fusion-autopilot run --mode incremental` (the daily delta-merge). Parses intent into scope flags (reusing the seed parser with --mode incremental), auto-satisfies preconditions, and runs the incremental-specific prechecks: a Fusion-PVO/bro
aidp-fusion-pipelineProductized Fusion → Oracle AI Data Platform pipeline with curated BICC extracts (GL/AR/AP/PO/Suppliers/Items), bronze/silver/gold medallion in Delta, conformed dimensions (account/calendar/org/supplier/item), gold marts (AR-Aging/AP-Aging/GL-Balance/PO-Backlog/Supplier-Spend), and MCP-native OAC wo
aidp-fusion-seedTurn a natural-language seed request into a correct, guarded `aidp-fusion-autopilot run --mode seed`. Parses intent into scope flags (--datasets / --layers / --strict-scope / --resume), auto-satisfies preconditions (validate + `/aidp-fusion-bootstrap` + cluster), guards the destructive replace-on-si
aidp-fusion-statusRead-only health + freshness report for the Fusion->AIDP pipeline. Answers 'did the last seed/incremental succeed?', 'what's stale or failed?', 'which marts are materialized?', 'what needs re-running?'. Cross-references fusion_autopilot_state (recorded run history) with the LIVE AIDP catalog (what's
aidp-restAIDP control-plane REST client primitives: OCI signing, workspace/cluster discovery, notebook upload via contents API, Job + JobRun lifecycle, taskRun output fetch + marker parsing. Use when a skill or tool needs to talk to a live AIDP cluster — uploading code, dispatching jobs, polling state, retri
aidpf-error-triageRead-only triage router for Oracle AIDP Fusion Autopilot failures. Use when the user pastes CLI output, a traceback, an `.aidp/diagnostics` path, a run id, or says bootstrap, validate, seed, incremental, dashboard, OAC, or workbook failed. Extract `AIDPF-*` codes, affected run id, dataset, layer, va
fusion-drift-doctorDiagnose and route Fusion->AIDP schema-drift / gate failures. When a run fails with a drift gate (AIDPF-2072 Fusion PVO drift, AIDPF-4070/4071 source-schema, AIDPF-2012 bronze fingerprint, AIDPF-4040 plan-hash), this read-only skill explains the code, probes the LIVE Fusion PVO, classifies each affe
incremental-mechanismChoose and verify the safest low-cost refresh mechanism for an Oracle Fusion AIDP bronze, silver, or gold node. Use when asked whether a node can be incremental, why a node re-extracts or rebuilds in incremental mode, how to reduce refresh cost, or when authoring a new node needs a refresh-strategy
mart-authorAuthor a new medallion node (gold mart, silver dim, or additive column) when the live gold layer AND the content pack both cannot serve a business need. Takes the user's business logic, inspects the Fusion PVO SOURCE schema (not bronze) for available raw fields, and authors the lowest-cost, additive
medallion-authorDraft a content-pack overlay extending the starter pack's variation-point candidate lists when `aidp-fusion-autopilot bootstrap` fails AIDPF-2010 / AIDPF-2011. Reads diagnostic artifacts under `.aidp/diagnostics/<run_id>/`, proposes new candidates from the observed bronze schema, presents a draft ov
oac-dataset-advisorTurn a natural-language dashboard request into a concrete OAC dataset plan, grounded in the LIVE AIDP gold layer. Inspects the actual materialized Delta tables in the AIDP catalog (evidence — never the content-pack YAMLs), checks what OAC datasets already exist via OAC MCP, then tells the operator e
oac-dataset-setupGuide the manual Oracle Analytics Cloud dataset checkpoint for the Fusion AIDP bundle. Consume an `oac-dataset-advisor` recommendation, produce exact OAC UI steps for creating or reusing the AIDP connection and dataset over live AIDP gold tables, pause while the operator creates the dataset, then ve
workbook-authoringAuthor regenerated OAC workbook JSON with deterministic template-first generation, runtime semantic validation checks, MCP save-validation, and preview iteration.