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/corridortech/posecap/ad-commitnpx skills add CorridorTech/PoseCap --skill ad-commitgit clone --depth 1 https://github.com/CorridorTech/PoseCapWhat 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 | $0.00118 | $0.01439 |
| Opus 5 | $0.00059 | $0.00720 |
| Sonnet 5 | $0.00024 | $0.00288 |
| Haiku 4.5 | $0.00012 | $0.00144 |
Grade A, and why
ad-commit 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<background_information>
Implements ADR-0023. Drafts atomic Conventional Commits with DCO Signed-off-by sign-off. Helper, not blocker — the senior engineer keeps decision authority.
Codex auto-trigger on description keywords is less mature than Claude Code's. If auto-invocation does not fire when the user mentions committing, staging, or signing off, invoke this skill manually. </background_information>
Route elsewhere when:
- The user wants to open a PR after committing → continue with
ad-prafter this skill finishes. - The user wants to merge an open PR →
ad-merge. - The diff is unfinished work that should not land yet → say so; do not commit speculatively.
Phase 1 — scope intake. Read the working tree:
git status --short
git diff --staged
git diff
Classify concerns. Group changes by Conventional Commits type (feat:, fix:, chore:, docs:, refactor:, test:, build:, ci:, perf:, style:, revert:), subsystem (directory / module), and causal relationship (a mechanical rename and its callers are one concern; a feat: and an unrelated fix: are two).
Decision:
- Single concern → proceed to Phase 3.
- Multiple concerns → present a numbered list and ask the user to confirm a stage-split plan. Example:
Detected 3 concerns:
1. feat(auth): add login flow (src/auth/, test/auth.test.js)
2. fix(cli): trim trailing newline (src/index.js)
3. docs(readme): refresh skill table (README.md)
Proceed concern-by-concern? (y/n, or "bundle them all into one commit")
When the user says "bundle them" or you are in genuine doubt, surface the call and proceed with their decision. Senior engineer keeps authority.
Phase 2 — stage-split (when needed). For each concern, in user-confirmed order:
- Stage exactly the files (or hunks) for this concern. Use
git add <path>for whole files;git add -p <path>when a hunk-level split is needed. Nevergit add -A/git add .when concerns are mixed. - Verify with
git diff --staged— only this concern's changes are present. - Draft + write the commit (Phases 3 + 4 below).
- Repeat for the next concern.
If a partial-file hunk split is too complex, surface the call: either accept the bundled commit for that file or ask the user to split the file edit first.
Phase 3 — draft message. Conventional Commits 1.0.0 format:
<type>(<scope>?): <subject>
<body — explains WHY when non-obvious; references task/ADR/issue>
Subject rules: ≤72 chars; imperative mood ("add", not "added"); no trailing period; lowercase after the colon; scope optional but recommended.
Body rules: explain why when non-obvious; reference task / ADR / issue (e.g. Closes task-0025, Per ADR-0023, Fixes #42); wrap at ~72 chars; blank line between paragraphs; no trailing summary of what the diff shows.
Type guide: feat: new user-visible capability; fix: bug fix; chore: maintenance; docs: documentation; refactor: no behavior change; test: tests only; build: build / packaging; ci: CI config; perf: performance; style: formatting; revert: reverts a prior commit.
Breaking changes: add ! after the type/scope and a BREAKING CHANGE: footer paragraph.
Phase 4 — sign + write. Resolve identity:
git config user.name
git config user.email
If either is unset, stop and ask the user to configure them. DCO sign-off requires real attribution; the skill will not invent it.
Append the trailer (always last line, blank line before it):
Signed-off-by: <Name> <email>
Do not add a Co-Authored-By: trailer (ADR-0023 §Decision 1 — DCO only).
Write the commit using a HEREDOC so multi-line body content is preserved verbatim:
git commit -m "$(cat <<'EOF'
<type>(<scope>): <subject>
<body paragraph 1>
<body paragraph 2>
Signed-off-by: <Name> <email>
EOF
)"
What ships with it
1 file 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.
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 · 123 lines · 118 tokens per session scan A 9df72977257d
ad-commit is a skill published in the GitHub repository CorridorTech/PoseCap (190 stars, last pushed 9d ago), licensed Apache-2.0. It adds 118 tokens to every session and 1,439 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…