role-technical-writer
Skill0 starsChanged 4 months agoMITView on GitHub
Use this skill for user-facing documentation, public API guides, release notes, migration notes, or documentation reviews where audience, structure, examples, or scope need deciding, even when the user only asks you to describe something. Routes to standards-documentation and standards-markdown instead of duplicating them.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- saemihemma/lead-producer-osstheir own version, first seen Sep 2, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- saemihemma/lead-producer-oss★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- mcpSkill
- pattern-orchestration-complexSkill
- pattern-task-breakdownSkill
- role-architectSkill
- role-code-reviewSkill
- role-data-engineerSkill
- role-developerSkill
- role-orchestratorSkill
- role-product-engineerSkill
- role-qa-engineerSkill
- role-security-auditorSkill
- role-site-reliability-engineerSkill
- standards-analysisSkill
- standards-api-designSkill
- standards-codeSkill
- standards-databaseSkill
- standards-documentationSkill
- standards-go-concurrencySkill
- standards-go-performanceSkill
- standards-goSkill
- standards-go-testingSkill
- standards-markdownSkill
- standards-observabilitySkill
- standards-pulumi-typescriptSkill
- standards-pythonSkill
- standards-securitySkill
- standards-shellSkill
- standards-skill-authoringSkill
- standards-testingSkill
- standards-typescriptSkill
- tool-storeSkill
- workflow-debuggingSkill
- workflow-pi-extensionSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Never: create a new summary or planning file unless the user asked for one
- Distinctive: line 41
Read the documentation type patterns when creating a README, API reference, code comment policy, architecture document, or procedural user …
- Distinctive: line 45
Start from the reader's goal, not the system's structure.
- Distinctive: line 46
Use concrete examples and show expected output.
- Distinctive: line 51
Check accuracy against the code and config before polishing prose.
- Distinctive: line 84
Do not create new summary files unless the user asks: Keep the summary in the conversation and update the existing docs instead.
- Distinctive: line 85
Do not document obvious code behavior: Document decisions and non-obvious constraints.
- Distinctive: line 87
Do not describe planned behavior as current.
- Distinctive: line 93
Before finishing documentation work:
- Distinctive: line 103
When the change documents a new or altered feature, also check the documents that sit outside the one you edited:
- Distinctive: line 120
With standards-api-design: document a public or partner contract accurately.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Provides: Audience identification, document type selection, drafting order, and review stance for documentation work.
Quick Reference
Goal: make the next reader action obvious
Principles: audience first, examples over abstraction, accurate over comprehensive
Never: create a new summary or planning file unless the user asked for one
Reference: skill:standards-documentation for structure and skill:standards-markdown for formatting
Workflow
1. Identify the audience
Each audience needs different content, and a document that serves all four usually serves none:
| Audience | Needs |
| End users | Goals, steps, expected results, troubleshooting |
| Developers | Contracts, examples, constraints, edge cases |
| Operators | Deploy, rollback, monitoring, escalation |
| Contributors | Setup, conventions, validation commands |
2. Choose the document type
- README: orientation and quick start.
- Guide: a task-based procedure.
- API docs: contract, examples, errors, compatibility.
- Spec or ADR: context, decision, alternatives, consequences.
- Migration notes: what changed, action required, rollback or fallback.
Read the documentation type patterns when creating a README, API reference, code comment policy, architecture document, or procedural user guide. Load only the section for the artifact requested.
3. Draft the minimum useful content
- Start from the reader's goal, not the system's structure.
- Use concrete examples and show expected output.
- Link to detail instead of embedding unrelated background.
4. Review
Check accuracy against the code and config before polishing prose. Verifying a command is worth more than improving a sentence. Then remove stale, redundant, or generic content.
Documentation Layout
Scale the structure to the project. A single accurate README.md beats a docs/ tree of stubs.
Small project:
README.md # Overview, quick start, installation
API.md # API reference
CONTRIBUTING.md # How to contributeLarge project:
docs/
README.md # Documentation home
getting-started/ # Installation, quick start
guides/ # Task-based procedures
reference/ # API and configuration
decisions/ # ADRsSplit a document only when readers with different goals keep colliding in it.
Gotchas
- Do not create new summary files unless the user asks. Keep the summary in the conversation and update the existing docs instead.
- Do not document obvious code behavior. Document decisions and non-obvious constraints.
- Do not let formatting work expand the scope of the requested change. Reformatting untouched sections buries the actual edit in diff noise.
- Do not describe planned behavior as current. Mark anything unverified.
Checklist
Before finishing documentation work:
- The target audience and the reader's next action are clear
- Examples, commands, links, and paths are accurate, or caveated as unverified
- The document is as short as its use case allows
- Markdown renders cleanly and code fences are balanced
- Stale or superseded content was removed, not just added around
- No new summary or planning file was created unless requested
- Formatting changes stayed within the requested scope
When the change documents a new or altered feature, also check the documents that sit outside the one you edited:
- README quick start, if the feature changes how someone gets started
- API reference, for new or changed endpoints and functions
- User guide, for how to actually use it
- Architecture docs, if the feature changes how the system fits together
- Changelog entry
- Migration note, if anything breaks for existing users
Integration with Other Skills
With standards-documentation: apply the skeleton and content conventions for the chosen type.
With standards-markdown: apply the formatting rules, including the SKILL.md carve-out.
With standards-api-design: document a public or partner contract accurately.
With role-product-engineer: turn a shipped feature into release notes and user-facing docs.
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.

