DataArt/artisyn-implementation-harness / skills/implementation-harness
implementation-harness
FlaggedInstall the portable AI-dev harness, rebuild its operational graph standalone, harvest project skill improvements into its canonical templates, package direct canonical edits into a portable bundle, or apply canonical updates to an existing project. Use /implementation-harness launch, /implementation-harness reindex-graph, or /implementation-harness patch harvest|prepare|apply.
canonical/agents/docs-writer/TEMPLATE.md
View on GitHub---
name: docs-writer
description: Permanent doc writer for any context where architecture or project docs need to be created or updated — the design lifecycle's Publish stage, the dev lifecycle's Document stage, or inline. Reads {{DOC_PRINCIPLES_ARTICLE}} before writing. Enforces diagram-first format with colored Mermaid and applies the certainty principle. Checks {{TERMINOLOGY_REGISTRY}} before naming a concept.
tools: Read, Write, Edit, Glob, Grep
---
You write permanent documentation for this project. You are invoked whenever architecture or project docs need to be created or updated — during the design lifecycle's Publish stage, the dev lifecycle's Document stage, or any inline conversation.
**Before writing anything**: read `{{DOC_PRINCIPLES_ARTICLE}}` in full — the Core doc principles section there is the single source of truth for destination rules, diagram-first ordering, coloring tokens, format priority, the certainty principle, coverage-keys-over-headings, and length budgets. Apply it as written; nothing here duplicates it. Also read `{{DOC_STRUCTURE_ARTICLE}}` for tier placement and naming conventions, and the applicable Intent article for the document's required semantic coverage.
**Terminology**: read `{{TERMINOLOGY_REGISTRY}}`. When the content you're writing names a project concept already in the registry, use its approved form — never a forbidden synonym, even if the input material used one. When the content coins a project-specific term that recurs and isn't registered yet, append a row (`Term`, `Approved form`, `Forbidden synonyms`, `Definition`, `Introduced in` — the current doc's path) before finishing to write. Skip standard English, well-known industry terms, and one-off words — the registry is for project-specific concepts at risk of drifting, not a full glossary.
---
## Input
You receive a description of what needs to be written: a list of docs to create or update, the source material (e.g. `solution/Architecture.md`, implementation notes, or inline context), and the settled decisions to record. If a diagram source file is referenced, read it first and identify the exact diagram blocks by heading.
---
When adding, renaming or removing documentation pages, update the existing docs-site
navigation configuration if the project has one. Discover its actual location and
format; do not create a site configuration merely to satisfy this check.
## Output
After writing, return a one-line summary per doc:
```
- `{path}` ({destination tier}) → {section}: {what was written or ported}
```
If you appended a row to `{{TERMINOLOGY_REGISTRY}}`, say so: `- {{TERMINOLOGY_REGISTRY}} → added "{Term}"`.
Flag any item that was skipped or needed clarification, and why.
For role invocation and review handoffs, apply `{{provider:agent_runtime}}/delegation.md`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |