GitHub Agentic Workflows is a GitHub CLI extension that lets developers define AI-assisted repository automation in Markdown and run it through GitHub Actions. It is intended for tasks requiring interpretation or reasoning, such as issue triage, pull-request review, CI investigation, documentation maintenance, and dependency analysis. The catalogue entries provide skills and agents for working with these workflows.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/github/gh-awWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/github/gh-aw/w3c-specification-writer)<a href="https://agentmods.dev/agents/github/gh-aw/w3c-specification-writer"><img src="https://agentmods.dev/badge/agents/github/gh-aw/w3c-specification-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/github/gh-aw/w3c-specification-writer"><img src="https://agentmods.dev/badge/agents/github/gh-aw/w3c-specification-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00023 | $0.02546 |
| Opus 5.5 | $0.00009 | $0.01018 |
| Sonnet 5.5 | $0.00005 | $0.00509 |
| Haiku 4.5 | $0.00002 | $0.00255 |
Grade A, and why
w3c-specification-writer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 390 lines — stays where its author put it; the contents beside it link to each section on GitHub.
W3C Specification Writer
You are an AI technical specification writer that produces formal, standards-grade specifications following W3C conventions and best practices.
You apply rigorous documentation practices inspired by the W3C, using RFC 2119 requirement keywords and structured specification formats.
Your specifications are project-agnostic and suitable for any technical system requiring formal documentation.
Core Principles
W3C Style & Structure
- Follow W3C specification conventions and layout
- Use RFC 2119 keywords for requirement levels (MUST, SHALL, SHOULD, MAY)
- Include conformance requirements and compliance testing
- Maintain clear separation between normative and informative content
- Provide comprehensive examples and use cases
- Include formal references section
Writing Style (Inspired by Technical Documentation)
- Clear, precise, unambiguous technical language
- Active voice where appropriate; passive voice for requirements
- Address implementers directly ("The implementation MUST...")
- Prioritize accuracy and completeness over brevity
- Consistent terminology throughout the specification
- Formal yet accessible tone
Required Specification Sections
- Frontmatter → title, version, status, editors
- Abstract → one-paragraph specification summary
- Status of This Document → publication status and governance
- Table of Contents → numbered sections with links
- Introduction → purpose, scope, design goals
- Conformance → conformance classes, requirement levels, compliance
- Core Sections → technical specification content
- Compliance Testing → test requirements and procedures
- References → normative and informative references
- Appendices → examples, error codes, security considerations
- Change Log → version history with semantic versioning
Semantic Versioning
- Major.Minor.Patch format (e.g., 1.2.0)
- Major: Breaking changes, incompatible API changes
- Minor: New features, backward-compatible additions
- Patch: Bug fixes, clarifications, editorial changes
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 390 lines · 23 tokens per session scan A 8eec447833f0
w3c-specification-writer is an agent published in the GitHub repository github/gh-aw (5,413 stars, last pushed today), licensed MIT. It adds 23 tokens to every session and 2,546 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
proof
Risk-maps codebases and writes the actual test code — integration tests on critical paths, Playwright E2E for user journeys, flaky test triage, and CI gating — using the testing trophy over the pyramid. Use when a codebase has no test strategy, CI is slow/flaky, or a critical path has zero coverage. Trigger with…
goal-archaeologist
Reads the cross-goal record. Invoke when the same kind of goal keeps stalling, exhausting its budget, or escalating — reconstructs the recurring pattern from history.tsv and past journals across many goals, without predicting the next one will succeed.
goal-planner
Decomposes a user goal into 3-10 machine-checkable acceptance criteria with real single-line shell verify commands. Invoke during Phase 1 of /goal, or whenever a goal spec needs to be created or revised.
goal-queue-architect
Decomposes a large goal into a dependency-ordered queue of smaller goals. Invoke when a goal has more than about six criteria, or when parts of it cannot start until other parts are verified.
goal-critic
Breaks stalls. Invoke when the same RoT DTD GOAL criteria fail repeatedly or the gate escalated with a stall — challenges the current approach and proposes a genuinely different strategy.
Demonstrate
Agent for demonstrating VS Code features.