New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Odc Agent Docs

by DouglasNeuroInformatics in DouglasNeuroInformatics/OpenDataCapture

Write or correct this repo's agent documentation — a workspace AGENTS.md (or its CLAUDE.md symlink), a playbook under .agents/docs/playbooks, an architecture doc, or the workspace map.

Apache-2.0Auto-check passedAgent Workflows

Install Odc Agent Docs

skills CLI
$ npx skills add DouglasNeuroInformatics/OpenDataCapture --skill odc-agent-docs -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install DouglasNeuroInformatics/OpenDataCapture odc-agent-docs --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/DouglasNeuroInformatics/OpenDataCapture.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/odc-agent-docs .claude/skills/odc-agent-docs && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
odc-agent-docs
GitHub stars
119
Token cost
~2.7k tokens
SKILL.md length
1,189 words
Files
2
Skills in repo
16
Repo updated
First seen
Licence
Apache-2.0

At a glance

Write or correct this repo's agent documentation — a workspace AGENTS.md (or its CLAUDE.md symlink), a playbook under .agents/docs/playbooks, an architecture doc, or the workspace map.

  • A documented rule is wrong
  • SKILL.md covers Find the home before you write, Where a fact belongs, House format — playbook and House format — AGENTS.md, plus 4 more sections
  • Calls pnpm and vitest
  • A change contradicts an AGENTS.md

What it does

Odc Agent Docs is an agent skill from DouglasNeuroInformatics/OpenDataCapture. Write or correct this repo's agent documentation — a workspace AGENTS.md (or its CLAUDE.md symlink), a playbook under .agents/docs/playbooks, an architecture doc, or the workspace map. Use when a documented rule is wrong or missing, when a change contradicts an AGENTS.md, or when another skill needs the house format for these files. Architectural decisions land in .agents/docs/architecture (no ADR flow here); investigating the outside world is research's job.

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Agent Workflows, covering Agent instruction files and Architecture decision records. The repository describes itself as: An electronic data capture platform for administering remote and in-person clinical instruments. The licence is Apache-2.0.

When your agent uses it

  • A documented rule is wrong
  • A change contradicts an AGENTS.md
  • Another skill needs the house format for these files

Example prompts

  • “/odc-agent-docs”

What it can do on your machine

Read from SKILL.md and the folder at commit 432dec8. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • pnpm
    • vitest

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use pnpm, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Odc Agent Docs loads about 2.7k tokens when it runs. Until then it costs about 120 tokens; SKILL.md has 1,189 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~120
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from DouglasNeuroInformatics/OpenDataCapture at commit 432dec8, republished under its Apache-2.0 licence (© DouglasNeuroInformatics). 1,189 words, ~2,699 tokens.

Download SKILL.mdSave it as .claude/skills/odc-agent-docs/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
odc-agent-docs
description
Write or correct this repo's agent documentation — a workspace AGENTS.md (or its CLAUDE.md symlink), a playbook under .agents/docs/playbooks, an architecture doc, or the workspace map. Use when a documented rule is wrong or missing, when a change contradicts an AGENTS.md, or when another skill needs the house format for these files. Architectural decisions land in .agents/docs/architecture (no ADR flow here); investigating the outside world is research's job.

No tool will tell you a fact landed in the wrong file — .agents/ is neither shipped nor linted (.agents/docs/workspace-map.md). A true sentence in the wrong document is silently correct, and the second copy is the one that goes stale: correct one, miss the other, and the repo now says two things.

Every fact has one home. Find it before you write it.

Find the home before you write

Grep for the fact, not for the file — a distinctive token from it, across docs and code:

sh
grep -rn "accessibleQuery(undefined" --exclude-dir=node_modules .
HitsWhat it meansWhat you do
nonethe fact is newplace it by the table below
onethat file is the homepoint at it by path — a second statement is the failure, not the service
two or moreyou have found a duplicationpick the home, cut the others to a one-line pointer, land it in the same change

Collapsing them is part of the change that found them, not a follow-up.

Where a fact belongs

The fact isHome
a rule that binds every directoryroot AGENTS.md
a convention or invariant of one workspacethat workspace's AGENTS.md
an ordered procedure where a skipped step fails silently.agents/docs/playbooks/<verb>-<rest>.md — verb first, kebab-case
structure spanning workspaces — a pipeline, a permission model.agents/docs/architecture/<topic>.md
what a workspace is, what it builds, who depends on it.agents/docs/workspace-map.md
the API or local status of a consumed @douglasneuroinformatics/*.agents/docs/packages/<name>.md
agent behaviour with a trigger of its owna skill — .agents/skills/writing-great-skills/SKILL.md decides its shape

A playbook ends in a ## Verify block. If you cannot write the commands that show the work landed, what you have is reference, and it belongs in an AGENTS.md or an architecture doc. The corollary: a playbook carries the order of operations and points outward for every rule.

House format — playbook

add-web-route.md is the template; read it before writing a new one.

  • Title, then where the conventions live — "Conventions live in apps/web/AGENTS.md; read it first. This file is only the ordering that matters." Line 3 says what the file is not.
  • The expected failure, before step 1. Say what a correct run looks like when it looks broken, or which single step is the one that fails silently — add-instrument.md names step 6 in its second sentence.
  • Numbered steps, each opening in bold with the action, and each carrying the consequence of skipping it. Tables inside a step where the step is a choice between shapes.
  • Name the file to copy. Every playbook points at a canonical example by path — logs.tsx, vendor/react@19.x, apps/api/src/groups/. add-vendor-package.md writes it as "Copy an existing wrapper rather than starting empty".
  • Close on ## Verify: a fenced sh block of real commands, plus a line on what a given failure means — add-web-route.md says a TS2345 there means only that the route tree is not regenerated yet.

The list heading is the one unfixed part: ## Steps, ## Checklist and going straight into the numbered list are all in use — match the playbook nearest your topic. ## Traps before ## Verify is optional and holds what no step owns (add-blog-post.md).

House format — AGENTS.md

  • Title is the workspace path — # apps/web, # packages/licenses.
  • Identity paragraph, a few lines: what it is, who consumes it, whether it builds.
  • Stakes in a blockquote where getting it wrong is expensive, as apps/api and packages/runtime-core do.
  • The trap comes before the conventions, under ## Traps or a heading that states the trap as a sentence (## Nothing in this directory is checked by anything, vendor/AGENTS.md).
  • Name the canonical file to read for each recurring task, by path, the way apps/api/AGENTS.md names src/groups/__tests__/groups.service.test.ts.
  • Close on how the work is checked — ## Tests giving pnpm exec vitest --project <name>, or stating there is no project and what covers the behaviour instead (packages/react-core/AGENTS.md). Where vitest is not the answer the section keeps another name (docs/AGENTS.md, ## Checking your work); apps/api closes on ## Build after it and cli/AGENTS.md has none.
  • Point down, not up. The root AGENTS.md already loads every turn — point outward only where the reader must leave, to the playbook for the procedure or the architecture doc for the model.

A new AGENTS.md needs a sibling CLAUDE.md symlink — ln -s AGENTS.md CLAUDE.md, committed (git mode 120000, one-line blob AGENTS.md). Every CLAUDE.md here is that symlink: never author or edit one, edit the AGENTS.md. Every tracked AGENTS.md has the symlink except those that are not this repo's conventions — the published packages/instrument-guidelines/AGENTS.md (root AGENTS.md explains why) and .agents/skills/vercel-react-best-practices/AGENTS.md, vendored.

Show full SKILL.md (464 more words)Show less

Prose rules

  • Bold lead-in on every rule, so a scan of the bold text is the summary.
  • State the failure, not the exhortation. "A story written anywhere else is silently not picked up" (storybook/AGENTS.md) does work that "be careful where you put stories" cannot.
  • Every non-obvious claim carries its path, and a decision carries the commit SHA that made it.
  • A number in prose goes stale; a table counts itself. runtime-and-vendor.md states one set of files as two different counts, prose against its own table. State a count only beside the command that reproduces it (workspace-map.md, pnpm ls -r --depth -1).
  • Write the fact, not the process that found it.

Tables that move together

Each row is one commit's worth of work.

You changedThese move with it
a workspace added, renamed or movedworkspace-map.md (its section table, Built vs source-only, Published to npm), root AGENTS.md "Where to look", the new AGENTS.md and its CLAUDE.md symlink
a vitest project added or removedtesting-strategy.md Vitest projects table, the count opening it and its "everything else has no unit tests" list; workspace-map.md's Vitest project column; that workspace's ## Tests section
a new playbookroot AGENTS.md "Step-by-step playbooks" row
a new architecture docroot AGENTS.md "Deeper reference, read on demand" row
a new .agents/docs/packages/<name>.md.agents/docs/packages/index.md (its section table's Docs link, the fidelity table) and the root AGENTS.md "Internal DNP packages" list
a new skillits .claude/skills/<name> symlink and agents/openai.yaml sidecar — below — and the skill index in .agents/skills/odc-orientation/SKILL.md

A skill nothing symlinks is a skill nothing loads. ln -s ../../.agents/skills/<name> .claude/skills/<name>, committed at git mode 120000, is what puts a skill in front of an agent; without it the SKILL.md is never read and nothing reports it. The sidecar is two keys — interface.display_name and interface.short_description; copy .agents/skills/review-pr/agents/openai.yaml.

Formatting is not your problem

There is nothing to make pass — the pre-commit hook formats rather than verifies (.agents/docs/playbooks/run-locally.md). No proseWrap is configured, so your line breaks survive: match the wrap of the file you are editing. .agents/skills is formatted; .claude/skills is in .prettierignore because its entries are symlinks, which are not formattable.

Done when

  • A grep for each fact you wrote returns one statement of it and any number of pointers.
  • Every path you cited resolves — ls each one.
  • Every bullet of the House format list for the kind you touched is satisfied — a playbook with its expected failure before step 1, its named file to copy and its ## Verify; an AGENTS.md with its path title, trap before conventions, and closing check — or your reply names the bullet you skipped.
  • Every table in the row you triggered above is updated, or your reply names the one you left and why.
  • Any workspace AGENTS.md your change contradicts is revised in the same commit, never left disagreeing (root AGENTS.md, "Before you are done").

© DouglasNeuroInformatics, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/odc-agent-docs of DouglasNeuroInformatics/OpenDataCapture.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 432dec8

Compare with similar skills

Odc Agent Docs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Odc Agent Docs compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Odc Agent Docs this skillDouglasNeuroInformatics/OpenDataCapture119—~2.7kAutomated safety check: PassApache-2.0
Living Docs Governanceqshanx/docs-governance132—~4.1kAutomated safety check: PassMIT
Ad Driftalexandremendoncaalvaro/CorridorKey-Runtime7551 repos~1.6kAutomated safety check: PassCustom licence
Authoring Claude Mdsammcj/agentic-coding162—~2.6kAutomated safety check: PassApache-2.0
Rememberantonio-orionus/Arroxy389—~571Automated safety check: PassMIT
Ad DriftCorridorTech/PoseCap224—~1.8kAutomated safety check: NotesApache-2.0

Similar skills

  • Living Docs Governance

    qshanx/docs-governance

    把长期项目的文档当成一个小系统来维护，防止文档腐烂——四份各司其职的脊柱文件（CLAUDE.md 共享章程 / CLAUDEMAP.md 地图 / PROJECTSTATUS.md 健康仪表盘 / PROJECTLOG.md 流水账）+ Codex 的 AGENTS.md 入口桥接 + 固定读序，并按需连接 ARCHITECTURE、CONTEXT、ADR、契约、测试、回归和 Issue…

    132 GitHub stars~4.1k tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed

  • Ad Drift

    alexandremendoncaalvaro/CorridorKey-Runtime

    Read-only drift audit — compare AGENTS.md, ARCHITECTURE.md, and ADR statuses against what the code actually does.

    755 GitHub starsUsed in 1 repo~1.6k tokens
    Agent WorkflowsAuto-check passed

  • Authoring Claude Md

    sammcj/agentic-coding

    Creating and maintaining CLAUDE.md project memory files and .claude/rules/ rule files that provide non-obvious codebase context.

    162 GitHub stars~2.6k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed

  • Remember

    antonio-orionus/Arroxy

    Persists a durable Arroxy lesson — a gotcha, user preference, workflow rule, or design decision — to the right tracked file (project memory, AGENTS.md, CONTEXT.md, dev-docs, or an ADR) so any coding…

    389 GitHub stars~571 tokensUpdated 3 days ago
    DevelopmentAuto-check passed

  • Ad Drift

    CorridorTech/PoseCap

    Read-only drift audit — compare AGENTS.md, ARCHITECTURE.md, ADR statuses, feature specs in doc/specs/, and documentation discipline against what the code actually does.

    224 GitHub stars~1.8k tokensUpdated yesterday
    DevelopmentAuto-check: notes

  • Documentation

    extra-org/extra

    Keep the repository's documentation accurate, honest, and synchronized with the code.

    113 GitHub stars~848 tokensUpdated 1 mo ago
    DevelopmentAuto-check passed

More from DouglasNeuroInformatics/OpenDataCapture

All 16 skills in this repo

  • Odc Testing

    DouglasNeuroInformatics/OpenDataCapture

    Test a change in Open Data Capture. An agent skill from DouglasNeuroInformatics/OpenDataCapture.

    119 GitHub stars~1.5k tokensUpdated yesterday
    Auto-check: notes

  • Review PR

    DouglasNeuroInformatics/OpenDataCapture

    Triage pull requests against upstream main — one verdict, action items, and whether it needs the user's eyes.

    119 GitHub stars~2.8k tokensUpdated yesterday
    Auto-check passed

  • Odc Open PR

    DouglasNeuroInformatics/OpenDataCapture

    Open a pull request against main for the current branch. An agent skill from DouglasNeuroInformatics/OpenDataCapture.

    119 GitHub stars~533 tokensUpdated yesterday
    Auto-check passed

  • Opening Issues

    DouglasNeuroInformatics/OpenDataCapture

    File a GitHub issue for Open Data Capture with its labels. An agent skill from DouglasNeuroInformatics/OpenDataCapture.

    119 GitHub stars~614 tokensUpdated yesterday
    Auto-check passed

  • Odc Debugging

    DouglasNeuroInformatics/OpenDataCapture

    What actually breaks in Open Data Capture — a failing build, lint, unit test or e2e run; a change with no visible effect; a suite that stays green after you broke the code on purpose; an error…

    119 GitHub stars~2.8k tokensUpdated yesterday
    Auto-check: notes

  • Odc Done

    DouglasNeuroInformatics/OpenDataCapture

    Close out a change to Open Data Capture before committing, opening a PR, or reporting it done.

    119 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Odc Agent Docs

What does Odc Agent Docs do?

Write or correct this repo's agent documentation — a workspace AGENTS.md (or its CLAUDE.md symlink), a playbook under .agents/docs/playbooks, an architecture doc, or the workspace map. Odc Agent Docs is an agent skill from DouglasNeuroInformatics/OpenDataCapture.agents/docs/playbooks, an architecture doc, or the workspace map.

When should I use Odc Agent Docs?

Odc Agent Docs fits situations like: A documented rule is wrong; A change contradicts an AGENTS.md; another skill needs the house format for these files.

How do I install Odc Agent Docs in Claude Code?

Run `npx skills add DouglasNeuroInformatics/OpenDataCapture --skill odc-agent-docs -a claude-code`. Or copy the skill folder (.agents/skills/odc-agent-docs in DouglasNeuroInformatics/OpenDataCapture) into .claude/skills/odc-agent-docs in your project. Claude Code loads it when a task matches its description.

How do I install Odc Agent Docs in Codex?

Run `npx skills add DouglasNeuroInformatics/OpenDataCapture --skill odc-agent-docs -a codex`. Or copy the skill folder (.agents/skills/odc-agent-docs in DouglasNeuroInformatics/OpenDataCapture) into .agents/skills/odc-agent-docs in your project. Codex loads it when a task matches its description.

Can I use Odc Agent Docs in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add DouglasNeuroInformatics/OpenDataCapture --skill odc-agent-docs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/odc-agent-docs, .gemini/skills/odc-agent-docs, .github/skills/odc-agent-docs and .opencode/skills/odc-agent-docs in your project.

What does Odc Agent Docs need to run?

Going by SKILL.md and its folder, Odc Agent Docs needs the command-line tools its instructions call (pnpm and vitest).

Does Odc Agent Docs access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Odc Agent Docs safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Odc Agent Docs use?

Odc Agent Docs is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Odc Agent Docs use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Odc Agent Docs?

Skills that share tags, products or a category with Odc Agent Docs: Living Docs Governance (qshanx/docs-governance, 132 stars), Ad Drift (alexandremendoncaalvaro/CorridorKey-Runtime, 755 stars), Authoring Claude Md (sammcj/agentic-coding, 162 stars) and Remember (antonio-orionus/Arroxy, 389 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Odc Agent Docs?

DouglasNeuroInformatics (a GitHub organization) maintains it in DouglasNeuroInformatics/OpenDataCapture, which has 119 GitHub stars. The repository holds 16 skills in this directory. The repository was last updated on October 7, 2026.

Source: DouglasNeuroInformatics/OpenDataCapture on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.