fabric-document
Skill49 starsChanged 6 months agoMITView on GitHub
Synthesizes all pipeline handoffs into a single human-readable project brief. Use when user says "generate docs", "create documentation", "document the project", or after validation passes. Do NOT use for deployment (use fabric-deploy) or architecture design (use fabric-design).
In this repository8 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- fabric-deploySkill
- fabric-designSkill
- fabric-discoverSkill
- fabric-healSkill
- fabric-testSkill
- grill-with-docsSkill
- handoffSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 22
Read these files in parallel from _projects/[name]/docs/ (all are independent inputs):
- Very specific: line 33
Follow the template in references/documentation-templates.md.
- Distinctive: line 52
Do NOT create README.md, architecture.md, deployment-log.md, or decisions/*.md
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
⚠️ Artifacts-only mode: When
deploy_modeisartifacts_only, this skill is NOT invoked — the pipeline runner handles this phase deterministically. The instructions below apply only whendeploy_modeislive.
Fabric Documentation
Goal: Produce ONE deliverable —
docs/project-brief.md— that a CTO can read in 10 minutes.
⚡ Pre-generated brief: The runner writes a structured
project-brief.mdsynthesized from all handoffs before this skill is invoked, with<!-- AGENT: FILL -->markers where narrative polish is needed. Your job: (1) verify facts against the source handoffs, (2) add narrative polish so the brief reads like a CTO memo rather than a template, (3) replace every<!-- AGENT: FILL -->marker.
Instructions
Step 1: Read Pipeline Handoffs
Read these files in parallel from _projects/[name]/docs/ (all are independent inputs):
discovery-brief.md— problem statement and signalsarchitecture-handoff.md— items, waves, decisions, trade-offs, diagramdeployment-handoff.md— what was deployed, config rationale, manual stepstest-plan.md— acceptance criteria, edge casesvalidation-report.md— pass/fail results
Step 2: Polish project-brief.md
Edit _projects/[name]/docs/project-brief.md in place (it is pre-generated).
Follow the template in references/documentation-templates.md. Key rules:
- ~1,500 words max — every fact appears exactly once
- Decisions inline — as a table in "Why This Architecture", NOT separate ADR files
- No separate files — do NOT create README.md, architecture.md, deployment-log.md, or decisions/*.md
- Synthesize, don't copy — distill the handoffs into narrative, don't paste YAML blocks
Step 3: Verify Completeness
The brief must answer these 5 questions:
- What problem are we solving? (from discovery)
- What did we build? (items + diagram from architecture)
- Why this approach? (decisions from architecture)
- How to deploy it? (from deployment handoff)
- Did it work? (from validation)
Constraints
- Documents only — no architecture decisions or deployments
- ONE output file:
project-brief.md - Do NOT create README.md, architecture.md, deployment-log.md, or decisions/*.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

