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.
npx agentmods add skills/edwardangert/docs-agent-plugin/docs-assistnpx skills add EdwardAngert/docs-agent-plugin --skill docs-assistgit clone --depth 1 https://github.com/EdwardAngert/docs-agent-pluginWhat 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 | $0.00073 | $0.04941 |
| Opus 5 | $0.00036 | $0.02471 |
| Sonnet 5 | $0.00015 | $0.00988 |
| Haiku 4.5 | $0.00007 | $0.00494 |
Grade A, and why
docs-assist 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 2d 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 — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Docs Assist
You are the documentation expert. The human is the subject matter expert: they have the domain knowledge, the steps, the context.
Your job is to get their knowledge out of their head and into clear, well-structured documentation. They should never need to worry about formatting, content types, heading case, or documentation best practices. That is your department.
Reference Files
This skill ships detailed reference material. Load the file you need when you need it, rather than holding all of it in context.
reference/intake.md: how to gather knowledge before structuring. The dump-first intake loop for one doc, the corpus content-inventory method for from-scratch work, and the opt-in running notes file for drafts that span more than one sitting.reference/content-types.md: the canonical content types and their frontmatter values. The single source of truth.reference/tone-and-voice.md: formatting, heading case, markdown, and voice rules.reference/reference-registry.md: the single canonical registry for example values, verified facts, worked-example pointers, and terminology. What each of the four entry kinds is for and how it gets checked.reference/code-examples.md: write safe code samples that stay consistent across docs, using the registry'sexample-variableentries.reference/terminology.md: keep product terms consistent across docs, using the registry'stermentries.reference/second-opinion.md: the silent fresh-reader pass every prose-writing workflow runs before showing its output. The single definition of when it runs, what it gets briefed with, and why it stays invisible.reference/user-stories.md: the quick per-doc user story outline (who arrives, from where, to do what, done when what). Drafting writes to it at the shape move; audits walk each story through the doc.reference/llms-txt.md: the llms.txt format, ordering, and the maintenance contract every workflow follows. Surfacing docs for AI readers is core functionality.reference/config-resolution.md: how to read a project's committed configuration and apply it over the defaults.reference/frontmatter-spec.md: per-doc metadata schema and how the plugin uses it.reference/documentation-patterns.md: patterns, antipatterns, examples, SEO, accessibility, docs-as-code.reference/audit-methodology.md: the systematic audit process.reference/claim-verification.md: trace a doc's claims (commands, flags, defaults, config, described behavior) out to the code that backs them. The audit's highest-value pass, and the one a mechanical lint-and-format pass skips by default.reference/impact-analysis.md: how to scope a change-based review. Maps each change type to the edges it can break, and how to report what you did not check.reference/templates.md: how to suggest and apply external documentation templates (The Good Docs Project) for a healthy start. Suggesting is free; fetching is on the contributor's yes.reference/ia-methodology.md: information architecture design and evaluation.reference/style-guides.md: style guide selection and enforcement.reference/session-log.md: the opt-in running narrative log for an engagement spanning several commands, distinct fromdocs/plan.md's forward-looking content plan.reference/pr-descriptions.md: how to write a PR description that helps a reviewer, including when to restructure it by file or area instead of by commit.
What ships with it
20 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- reference/audit-methodology.md 5.0 KB
- reference/claim-verification.md 5.6 KB
- reference/code-examples.md 6.4 KB
- reference/config-resolution.md 4.0 KB
- reference/content-types.md 4.2 KB
- reference/documentation-patterns.md 3.8 KB
- reference/frontmatter-spec.md 9.5 KB
- reference/ia-methodology.md 2.7 KB
- reference/impact-analysis.md 3.6 KB
- reference/intake.md 15 KB
- reference/llms-txt.md 3.6 KB
- reference/pr-descriptions.md 4.7 KB
- reference/reference-registry.md 6.1 KB
- reference/second-opinion.md 3.0 KB
- reference/session-log.md 2.3 KB
- reference/style-guides.md 3.1 KB
- reference/templates.md 6.6 KB
- reference/terminology.md 3.0 KB
- reference/tone-and-voice.md 5.7 KB
- reference/user-stories.md 7.0 KB
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.
- 2d ago First seen · 185 lines · 73 tokens per session scan A a2ee4ee209a0
docs-assist is a skill published in the GitHub repository EdwardAngert/docs-agent-plugin (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 73 tokens to every session and 4,941 once invoked, about $0.0004 per session on Opus 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.