gitadmin/skillfactor-pipeline
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
quality/reference-standard
skillfactor-pipeline/adapters/openai/SPEC.md
skillfactor-pipeline 7a72e45b93 feat(arch): openai/gemini adapter scaffolds + MCP gateway + M365 mining specs 
Same canonical content, three thin adapters; FastMCP scaffold with the
five gateway tools (layer-1 guard enforced by the lesson schema); M365
spec with assumed minimal DDL, read-only queries, extraction-time
anonymization and a review-queue with no auto-merge.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PDKeXvpT6tENSvyQGLV1Uq
2026-07-10 05:42:21 +02:00

1.3 KiB
Raw Permalink Blame History

ChatGPT adapter — spec

Build target from the same canonical content (no content changes).

Bundle per profession (build_openai.py → dist/<slug>/)

  • instructions.md — system instruction generated from the canonical profession content: role framing, retrieval cascade note, and the FULL §3 contribution behaviour (switch, full-text confirmation, duplicate check, anonymization, PR-only) including the canonical trigger prompt. Fits the 8k-char Custom-GPT instruction limit (truncates competence lists, links the knowledge files instead).
  • knowledge/ — the competence files + knowledge.md as upload files for the GPT's knowledge store (retrieval).

Integration (manual, per Custom GPT)

  1. Create a Custom GPT; paste instructions.md as Instructions.
  2. Upload knowledge/* as Knowledge.
  3. Write path: add an Action from openapi.yaml (this folder) — it fronts the SkillFactor MCP gateway's HTTP facade (services/skillfactor-mcp, endpoint /actions/propose_lesson etc.). Auth: API key header (SKILLFACTOR_TOKEN), no key material in this repo (.env.example only).

Status

  • Generator scaffold: RUNNABLE (build_openai.py).
  • openapi.yaml: complete spec for propose_lesson/get_profession_context.
  • The gateway HTTP facade itself is spec-level (see services/…/SPEC.md).