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-prnpx skills add CorridorTech/PoseCap --skill ad-prgit 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 | $0.00110 | $0.01631 |
| Opus 5 | $0.00055 | $0.00816 |
| Sonnet 5 | $0.00022 | $0.00326 |
| Haiku 4.5 | $0.00011 | $0.00163 |
Grade C, and why
ad-pr scanned grade C with 1 finding 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
Matrix awareness: if CI runs a multi-version matrix and local runs one version, note the gap but do not refuse — matrix mirroring belongs to ad-hooks. Refuse only on outright red. How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<background_information>
Implements ADR-0024 and ADR-0032 (doc/adr/0032-ci-failure-is-local-gate-gap.md). Opens a PR via gh pr create with a uniform body shape (Summary / Test plan / Links). Helper posture on scope, links, and body drafting — warnings surface without refusing. Hard gate on local quality: the skill refuses to open a PR when pre-push / CI-mirror gates exit non-zero (WORKFLOW §11 — CI failure is a local gate gap). No --no-verify symmetric bypass; users who need to open a red draft invoke gh pr create --draft directly.
Codex auto-trigger on description keywords is less mature than Claude Code's. If auto-invocation does not fire when the user mentions opening a PR or submitting changes for review, invoke this skill manually. </background_information>
Route elsewhere when:
- The branch has uncommitted or unsplit work →
ad-commitfirst. - The user wants to merge an already-open PR →
ad-merge.
Phase 1 — preflight. Check gh is installed and authenticated:
gh --version
gh auth status
If gh is missing, surface: "gh CLI not installed. Install: https://cli.github.com/ then run gh auth login. This skill cannot open the PR without it." Soft-fail — do not silently fall back to git push only.
If gh auth status fails, surface the same hint with gh auth login.
Check the branch is pushed:
git rev-parse --abbrev-ref @{u}
git rev-list --count <base>..HEAD
If no upstream is set or the branch is unpushed, prompt: "Branch not pushed to origin. Run git push -u origin <branch> first? (y/n)". On confirm, run the push and continue.
Detect the base branch:
gh repo view --json defaultBranchRef --jq .defaultBranchRef.name
Default to main if the call fails.
Run local gates before opening the PR (WORKFLOW §11 — CI failure is a local gate gap). The gate check runs after git push has landed the branch (so the pre-push hook has already fired once) and before gh pr create — the goal is to catch the case where a developer skipped the pre-push hook, ran on a different matrix leg than CI, or wired the runner incompletely, and to refuse to open the PR against a red tree:
- Detect the pre-push tier — read
lefthook.yml,.husky/pre-push,.pre-commit-config.yaml(pre-push stage), or.git/hooks/pre-push. - Run the detected commands explicitly. If no hook runner is detected, fall back to reading
.github/workflows/*.yml(or the detected CI surface) and run its test / lint / typecheck / build commands locally. - Refuse on red. If any gate exits non-zero, surface: "Pre-push gate failed: . Fix locally before opening the PR — pushing red-CI diffs burns cloud minutes. Rerun ad-pr after fixing." Do not offer
--no-verifyor a bypass. WORKFLOW §11 binding. - Warn on absent gates. If no hook runner and no CI config exist, surface: "No pre-push or CI config detected — opening PR without local gate check. Wire ad-hooks for coverage." Continue.
Matrix awareness: if CI runs a multi-version matrix and local runs one version, note the gap but do not refuse — matrix mirroring belongs to ad-hooks. Refuse only on outright red.
Phase 2 — scope assembly. Gather context:
git log <base>..HEAD --pretty='%h %s'
git diff --stat <base>...HEAD
Infer the PR title:
- Dominant Conventional Commits type across
<base>..HEAD. If one commit, use its type / scope / subject directly. If multiple commits and one type dominates, use that. If mixed, surface to the user: "Branch mixesfeat:+fix:. Use which type for the PR title?" - Scope = most common scope across the commits (or empty if inconsistent).
- Subject synthesized from the changes, ≤70 chars, imperative.
Detect back-links:
- Scan commit message bodies for
task-NNNN,ADR-NNNN,spec-NNNN,#<issue>,Closes,Fixes,Per,See. - Scan changed file paths for
doc/tasks/NNNN-*.md,doc/adr/NNNN-*.md,doc/specs/NNNN-*.md.
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 · 124 lines · 110 tokens per session scan C e6bd803887c2
ad-pr is a skill published in the GitHub repository CorridorTech/PoseCap (190 stars, last pushed 10d ago), licensed Apache-2.0. It adds 110 tokens to every session and 1,631 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). 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.
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.
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.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…