Promote knowledge for Microsoft review skills (#153)

Move canonical knowledge for Microsoft-owned review domains into the Microsoft layer and document the skill/knowledge co-location policy.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Co-authored-by: Jesper Schulz-Wedde <jesper.schulzwedde@microsoft.com>
Copilot-Session: 2a6ea875-d38e-4f30-aadb-0d606f9be231
This commit is contained in:
Jesper Schulz-Wedde 2026-09-03 15:06:01 +02:00 • committed by GitHub
parent bca8f478d8
commit 4f0a13a801
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
88 changed files with 11 additions and 7 deletions

View file

@ -58,7 +58,7 @@ Each action skill is a markdown file that specifies what to do at each step. The
| **Worklist** | Narrow from N candidates to the M that apply to this specific task. |
| **Action** | Apply the relevant knowledge and produce structured output. |
Example: a performance review skill sources from `/microsoft/knowledge/performance/` and `/community/knowledge/performance/`, filters to `bc-version: 26` and `technologies: [al]`, narrows the 25 candidate files to the 8 that apply to the 15 objects changed in the PR, and then evaluates each file against the diff.
Example: the Microsoft-owned performance review skill selects `performance` entries across every enabled layer, filters to `bc-version: 26` and `technologies: [al]`, narrows the candidate files to those that apply to the changed objects, and then evaluates each file against the diff. Its canonical corpus lives beside it under `/microsoft/knowledge/performance/`; cross-layer entries are limited to custom overrides or short-lived promotion work.
At this point the agent reads READ and DO on demand — it needs READ to interpret each knowledge file's frontmatter and sections, and DO to shape its output. Those contracts are fetched when first needed, not as part of bootstrap.
@ -104,7 +104,7 @@ Orchestrators MUST tolerate an absent `domain` in reports from older producers.
into Entry's task context. Entry and the dispatched action skills remain
authoritative.
- **Layers decide authority, not code.** The agent sees `/microsoft/` and `/community/` together; if two files conflict, the precedence rule defined in READ resolves it. A partner fork can disable `/community/` — that's a config choice, not a code change.
- **Knowledge and skills evolve independently.** A new knowledge file requires no skill changes — existing skills pick it up via frontmatter filters. A new skill requires no knowledge changes — it sources from what's already there.
- **Knowledge and skills evolve independently within their owning layer.** A new knowledge file requires no skill changes because existing skills pick it up via frontmatter filters. Layer placement still follows skill ownership, so promoting a skill also promotes its canonical corpus.
## The mental model, in one sentence