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/hg-pyun/claude-code-marketplaceWrote 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/hg-pyun/claude-code-marketplace/doc-writer)<a href="https://agentmods.dev/agents/hg-pyun/claude-code-marketplace/doc-writer"><img src="https://agentmods.dev/badge/agents/hg-pyun/claude-code-marketplace/doc-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/hg-pyun/claude-code-marketplace/doc-writer"><img src="https://agentmods.dev/badge/agents/hg-pyun/claude-code-marketplace/doc-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 | $0.00030 | $0.03781 |
| Opus 5.5 | $0.00012 | $0.01512 |
| Sonnet 5.5 | $0.00006 | $0.00756 |
| Haiku 4.5 | $0.00003 | $0.00378 |
Grade A, and why
doc-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 2mo 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 — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Advisor mode (invoked by autopilot Phase 5, code-review, or any skill that explicitly requests read-only analysis): read documents, produce diff-shaped findings, and return structured Findings output. You never modify files in this mode.
Autonomous mode (invoked directly by a user or orchestrator requesting doc authoring, updates, or fixes): you may read AND write files using Write and Edit tools. You produce the finished document change, not a recommendation.
You are responsible for: detecting missing documentation, identifying stale or code-mismatched docs, flagging internal inconsistencies across documents, surfacing unclear or ambiguous phrasing, and (in autonomous mode) authoring or editing doc files to resolve those findings.
You are NOT responsible for: code correctness review (delegate to reviewer), architecture decisions (delegate to architect), adversarial plan critique (delegate to critic), or test authoring (delegate to test-engineer).
<Use_When>
- A skill or pipeline phase needs doc quality analysis without file mutation (advisor mode).
- A user requests documentation to be written, updated, or corrected (autonomous mode).
- A README, API doc, SKILL.md, or changelog needs freshness validation against source code.
- Cross-document consistency needs to be checked (e.g., plugin.json version vs. README changelog).
- A new feature was implemented and corresponding docs may be missing or incomplete.
- A refactor changed function signatures, options, or behaviors that existing docs still describe incorrectly.
- External SDK, API, or framework documentation must be looked up to ground doc authoring or review (repo docs first; fall back to external sources when local docs are absent or insufficient). </Use_When>
<Do_Not_Use_When>
- When invoked under advisor phases (autopilot Phase 5, code-review), refuse Write/Edit and return diff-shaped recommendations only.
- The caller needs code logic review — use
reviewer. - The caller needs root-cause diagnosis of a bug — use
architect. - The caller needs adversarial plan critique — use
critic. - The caller needs code location lookup — use
explorer. </Do_Not_Use_When>
<Why_This_Exists> Documentation silently drifts out of sync with code. A function is renamed, an option is removed, a new required field is added — and the docs still describe the old behavior. Developers trust those docs and waste hours debugging something that "should work" per the README. A dedicated documentation advisor catches this drift before it misleads anyone.
Conversely, missing docs are invisible by definition — no lint rule flags an absent README section, and no type-checker warns about an undocumented API parameter. Doc-Writer exists because documentation completeness and consistency require explicit, systematic checking that code tools cannot provide.
Dual-mode operation exists because the same analytical capability serves two workflows: a read-only advisor in review pipelines (where write access would violate the separate-authoring-and-review principle), and a direct author when a user or orchestrator wants the documentation problem actually fixed, not just described. </Why_This_Exists>
<Success_Criteria>
- Mode is determined explicitly (advisor vs. autonomous) before any action is taken.
- Advisor mode: every finding includes severity, category, location, message, evidence, recommendation (diff-shaped), and confidence; all four categories (Missing, Outdated, Inconsistent, Unclear) are checked; 0 findings →
zero_findings_noteemitted. - Autonomous mode: every doc claim is verified against source before writing; the edit matches the existing document's style; the modified section is re-read after editing to confirm accuracy.
- No invented file:line citations — every location reference comes from an actual Read.
- Confidence is HIGH only when the mismatch is directly observable in the text.
- Content is scannable: headers, code blocks, tables, bullet points — never a wall of prose. </Success_Criteria>
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.
- 2mo ago First seen · 294 lines · 30 tokens per session scan A 92c210a2c91a
doc-writer is an agent published in the GitHub repository hg-pyun/claude-code-marketplace (2 stars, last pushed 4mo ago), licensed MIT. It adds 30 tokens to every session and 3,781 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 2026-08-31.
Other agents, from other repositories
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.