grafel-tech-docs
Generate the per-repo and group-level technical documentation set — overview, modules, reference, cross-cutting, group synthesis, cross-links, and patterns.
Skill by cajasmota
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/cajasmota-grafel-tech-docs
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/grafel-tech-docs
What's inside
- Skill name
grafel-tech-docs- Description
- Generate the per-repo and group-level technical documentation set — overview, modules, reference, cross-cutting, group synthesis, cross-links, and patterns. For engineers. Runs 13 orchestrated passes using the grafel MCP as the primary navigation surface. Supports delta re-runs (only rewrite modules whose graph changed).
- License
- MIT
- Instructions
- 137 lines in SKILL.md · view on GitHub
Files (58)
- SKILL.md
- _graph-searchability.md · conventions/
- django.md · conventions/
- fastapi.md · conventions/
- generic.md · conventions/
- go-stdlib.md · conventions/
- infra-cdk.md · conventions/
- infra-k8s.md · conventions/
- infra-terraform.md · conventions/
- nodejs-generic.md · conventions/
- python-generic.md · conventions/
- react-native.md · conventions/
- react.md · conventions/
- vite.md · conventions/
- http_endpoint.md · examples/
- message_topic.md · examples/
- process_flow.md · examples/
- api.md · output-templates/
- business-capability.md · output-templates/
- business-domain-glossary.md · output-templates/
- business-journey.md · output-templates/
- business-overview.md · output-templates/
- business-rules.md · output-templates/
- flows.md · output-templates/
- glossary.md · output-templates/
- group-synthesis.md · output-templates/
- how-to-local-dev.md · output-templates/
- module-readme.md · output-templates/
- overview.md · output-templates/
- reference-config.md · output-templates/
- reference-dependencies.md · output-templates/
- reference-deployment.md · output-templates/
- reference-misc.md · output-templates/
- reference-scripts.md · output-templates/
- 00-domain-qa.md · prompts/
- 01-inventory.md · prompts/
- 02-plan.md · prompts/
- 03-overview.md · prompts/
- 03a-generation-time-repair.md · prompts/
- 04-cluster.md · prompts/
- 05-reference.md · prompts/
- 06-cross-cutting.md · prompts/
- 07-group-synthesis.md · prompts/
- 08-cross-link.md · prompts/
- 10-pattern-convergence.md · prompts/
- 11-pattern-cross-link.md · prompts/
- 12-pattern-prose.md · prompts/
- 20-llm-orchestrate.md · prompts/
- anchor-contract.md · snippets/
- business-voice.md · snippets/
- cross-link-format.md · snippets/
- docgen-repair-emission.md · snippets/
- link-hygiene.md · snippets/
- progress-tracking.md · snippets/
- verification-checklist.md · snippets/
- http_endpoint.md · templates/
- message_topic.md · templates/
- process_flow.md · templates/
SKILL.md (first lines)
# grafel-tech-docs
Generate the complete technical documentation set for a registered grafel group: per-module READMEs, API reference, cross-cutting concerns, group synthesis, cross-repo links, and pattern library.
This skill is a direct extraction of the technical tier (Passes 0–12 + Pass 20) from `/generate-docs`. It is the largest and most expensive skill in the grafel family.
## CRITICAL TOOL DISCIPLINE (enforced on every pass — read before ANY action)
For ANY question about "what entities/files exist in this codebase", "who calls X", "what does Y import", "what's in module Z", you MUST use grafel MCP tools: `grafel_inspect`, `grafel_find`, `grafel_subgraph`, `grafel_orient (view=overview)`, `grafel_orient (view=clusters)`, `grafel_trace`, `grafel_orient (view=me)`.
You are STRICTLY FORBIDDEN from using `find`/`ls`/`wc`/`grep` on the codebase **for entity discovery**. Use Bash ONLY for reading specific source line ranges that `grafel_get_source` returns, running `grafel docgen --llm-mode=apply`, and writing output files into the staging directory.
If the MCP returns empty or seems wrong, file a side ticket and ABORT — do NOT silently substitute grep results for graph queries.
### Pre-flight assertion (FIRST action in every pass)
Call `grafel_orient (view=me)` before doing anything else. If it errors: ABORT with "grafel MCP not configured for this directory. Run `/mcp` to fix, then re-invoke `/grafel-tech-docs`."
## When to use this skill
Invoke when the user asks for any of:
- "Document this repo / this group."
- "Regenerate the docs after the recent refactor."
- "Write API reference / module guide / cross-repo overview."
- "Fill the LLM bundle" / "run docgen in LLM mode" / "orchestrate the bundle."
Do not invoke it for business-tier docs (that is `/grafel-business-docs`), enrichment (that is `/grafel-graph-enrich`), or residual cleanup (that is `/grafel-resolve`).
## Prerequisites
- A running grafel daemon (`grafel status` should show "running").
- A resolved grafel group.
- Per-repo `<repo>/.grafel/graph.json` from the most recent index run.
- Recommended (not required): run `/grafel-resolve` first to minimize residual edges. Run `/grafel-graph-quality` to confirm the graph foundation is solid before spending tokens on writing.
## Staging-dir + atomic promote architecture
Documentation is written into a **staging directory** during the run and atomically promoted to the canonical store only when all passes complete without errors. This makes in-repo write accidents architecturally impossible.
1. **Pass 2** calls `grafel_docgen(action="start", group="<group>")` → receives `run_id` and `staging_path`.
2. **Passes 3–12** write doc files into `<staging_path>/<relative-path>` using the Write tool.
3. **Pass 8** and **Pass 14** (if enrichment exists) call `grafel_docgen(action="validate", run_id)`.
4. **Pass 20** (or end of pass chain) calls `grafel_docgen(action="promote", run_id)` to atomically move staging → canonical.
## Pass chain
| Pass | Prompt | Purpose | Est. time |
|------|--------|---------|-----------|
| 0 | `prompts/00-domain-qa.md` | First-run domain interview | 5–10 min interactive |
| 1 | `prompts/01-inventory.md` | Discover repos and entities | 2–5 min |
| 2 | `prompts/02-plan.md` | Per-module documentation plan with token estimates | 2–3 min |
| 3 | `prompts/03-overview.md` | Repo-level overview.md for every repo | 3–5 min |
| 3a | `prompts/03a-generation-time-repair.md` | Inline repair hook for Passes 3–6 + 12 | (integrated) |
| 4 | `prompts/04-cluster.md` | Per-module deep-dive (parallel writer subagents) | 5–20 min |
| 5 | `prompts/05-reference.md` | Reference docs: API, config, deployment | 3–8 min |
| 6 | `prompts/06-cross-cutting.md` | Cross-cutting: auth, logging, error handling | 2–5 min |
| 7 | `prompts/07-group-synthesis.md` | Group-level synthesis page | 3–5 min |
| 8 | `prompts/08-cross-link.md` | Validate links + cross-repo link candidates | 2–4 min |
| 10 | `prompts/10-pattern-convergence.md` | Aggregate pattern candidates | 2–3 min |
| 11 | `prompts/11-pattern-cross-link.md` | Populate pattern documentation_url fields | 1–2 min |
| 12 | `prompts/12-pattern-prose.md` | Emit docs/patterns/<category>/<id>.md | 2–4 min |
| 20 | `prompts/20-llm-orchestrate.md` | LLM-mode orchestrate (standalone; only when bundle files exist) | varies |
**Note:** Passes 1a and 1b (residual repair) are NOT part of this skill. Run `/grafel-resolve` before invoking this skill. Pass 3a is an inline hook used by writer passes, not a standalone pass.
**Total wall time:** 25–65 min (small repos), 1–4 h (large repos). Pass 4 is the critical path; it parallelises across module clusters.
## LLM-mode orchestrate (Pass 20)
Pass 20 is a standalone mode triggered only when `*-page-bundle.json` files exist in the docgen output directory. Do NOT use Pass 20 inside the standard Passes 0–12 pipeline. See `prompts/20-llm-orchestrate.md` for the full procedure.
## Conventions
The skill applies stack-specific conventions to every writer subagent. See `conventions/` for registered conventions. Every convention requires `conventions/_graph-searchability.md` first.
If the agent encounters a stack with no matching convention, stop and direct the user to run the `extend-convention` skill.
## Quality gates
Before any pass commits output, the writer runs `snippets/verification-checklist.md`. The checklist enforces:
- **`snippets/anchor-contract.md`** — emitted anchors derived from actual headings (never hand-authored).
- **`snippets/link-hygiene.md`** — link targets are real generated doc files; no in-source-tree links.
- **Volume control** (`prompts/02-plan.md` § Step 1b) — merge thin modules, cap by LOC, no empty stub pages.
## Docgen Repair Feedback Contract
Writer passes (3, 3a, 4, 5, 6, 12) emit repair candidates to `docgen-repairs.jsonl` when they discover facts the static extractor missed. See `snippets/docgen-repair-emission.md` for the full emission procedure.
## Output layout
```
~/.grafel/docs/<group>/<repo-slug>/
overview.md
modules/<module-slug>/
README.md
api.md
flows.md
reference/
config.md
deployment.md
scripts.md
dependencies.md
how-to/local-dev.md
glossary.md
~/.grafel/docs/<group>/
group-synthesis.md
cross-links.md
docs/patterns/<category>/<id>.md
```
## grafel MCP tool surface
- `grafel_orient (view=me)`, `grafel_find`, `grafel_inspect`, `grafel_subgraph`
- `grafel_trace`, `grafel_trace`, `grafel_orient (view=clusters)`, `grafel_orient (view=overview)`
- `grafel_get_source`, `grafel_index_status`
- `grafel_findings (action=save)`, `grafel_findings (action=list)`
- `grafel_cross_links`, `grafel_docgen_apply (kind=enrichments)`, `grafel_docgen_apply (kind=repairs)`
- `grafel_patterns`From cajasmota/grafel, MIT.
More from cajasmota/grafel
- 0.607score
grafel-feedback
cajasmota
Generate a privacy-preserving anonymized quality report for sharing with grafel maintainers, then layer an agent-driven "Findings & Interpretation" section on…
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-graph-quality
cajasmota
Continuous health check of the grafel knowledge graph.
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-security-audit
cajasmota
Two-phase security audit of the indexed group.
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-business-docs
cajasmota
Generate group-synthesised, PM-facing documentation — product capabilities, business domain glossary, user journeys as plain-language narratives, business…
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-resolve
cajasmota
Surface and resolve residual edges — runtime dispatch, dynamic URLs, ambiguous bindings — so the graph is fit-for-purpose before any downstream skill consumes…
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-test-page
cajasmota
Test the grafel docgen LLM-mode loop end-to-end on a single entity.
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
using-grafel
cajasmota
Teaches an AI agent how to use grafel MCP tools effectively when working on a registered codebase.
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-consult
cajasmota
Hire a specialist consultant (architect, security-auditor, business-analyst, performance-reviewer, refactor-critic, api-designer, data-engineer, qa-reviewer)…
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-graph-enrich
cajasmota
Annotate the graph with structured YAML frontmatter on http_endpoint, process_flow, and message_topic entities so the dashboard panels (Paths, Flows…
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-help
cajasmota
Overview of the grafel skill family.
SkillBFreshness: B★ 24Claude Code +9 - 0.601score
grafel-graph-read
cajasmota
Shared grafel read protocol — status → inspect → expand.
SkillBFreshness: B★ 24Claude Code +9 - 0.599score
grafel-graph-write
cajasmota
Shared grafel persistence protocol — save findings only on explicit user request.
SkillBFreshness: B★ 24Claude Code +9
Signals
- Stars
- 24
- Forks
- 7
- Open issues
- 282
- Last commit (this folder)
- Jun 24, 2026 (4 months ago)
- Last release
- Sep 23, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/cajasmota/grafel
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from cajasmota
- 0.607score
grafel-graph-quality
cajasmota
Continuous health check of the grafel knowledge graph.
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-security-audit
cajasmota
Two-phase security audit of the indexed group.
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-business-docs
cajasmota
Generate group-synthesised, PM-facing documentation — product capabilities, business domain glossary, user journeys as plain-language narratives, business…
SkillBFreshness: B★ 24Claude Code +9 - 0.607score
grafel-feedback
cajasmota
Generate a privacy-preserving anonymized quality report for sharing with grafel maintainers, then layer an agent-driven "Findings & Interpretation" section on…
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-test-page
cajasmota
Test the grafel docgen LLM-mode loop end-to-end on a single entity.
SkillBFreshness: B★ 24Claude Code +9 - 0.604score
grafel-graph-enrich
cajasmota
Annotate the graph with structured YAML frontmatter on http_endpoint, process_flow, and message_topic entities so the dashboard panels (Paths, Flows…
SkillBFreshness: B★ 24Claude Code +9
Cite this page
grafel-tech-docs skill. Agent Mart. https://agentmart.dev/skills/cajasmota-grafel-tech-docs (last updated 2026-10-24)