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-mergenpx skills add CorridorTech/PoseCap --skill ad-mergegit clone --depth 1 https://github.com/CorridorTech/PoseCapWrote 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/skills/corridortech/posecap/ad-merge)<a href="https://agentmods.dev/skills/corridortech/posecap/ad-merge"><img src="https://agentmods.dev/badge/skills/corridortech/posecap/ad-merge.svg" alt="Measured on agentmods" 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 | $0.00134 | $0.01306 |
| Opus 5 | $0.00067 | $0.00653 |
| Sonnet 5 | $0.00027 | $0.00261 |
| Haiku 4.5 | $0.00013 | $0.00131 |
Grade A, and why
ad-merge 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 4d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<background_information>
Implements ADR-0025. Evaluates a PR's mergeability and performs the merge via gh pr merge. CI green is the only hard gate; everything else surfaces as a warning the senior engineer decides on.
Codex auto-trigger on description keywords is less mature than Claude Code's. If auto-invocation does not fire when the user mentions merging a PR or evaluating mergeability, invoke this skill manually. </background_information>
Route elsewhere when:
- No PR exists yet →
ad-prfirst. - Commits are not on the branch yet →
ad-commitfirst.
Phase 1 — preflight. Check gh is installed and authenticated:
gh --version
gh auth status
If absent or not authed, surface the install / gh auth login hint and stop (same soft-fail rule as ad-pr).
Resolve the target PR:
- If the user passed a PR number / URL, use it.
- Else
gh pr view --json number,headRefName,baseRefNameagainst the current branch. If that fails, surface: "No PR found for branch<name>. Open one withad-prfirst."
Phase 2 — evaluate. Run the structured check and report each line:
gh pr checks <num>
gh pr view <num> --json mergeable,mergeStateStatus,reviews,number,title,headRefName,baseRefName
gh api repos/:owner/:repo/pulls/<num>/comments
Findings format (pass / warn / fail):
CI status: <pass | pending | fail>
Fresh-context review: <pass | warn — none found>
Linked task / ADR: <pass | warn — none>
Unresolved comments: <pass | warn — N unresolved>
Mergeability: <pass | dirty | blocked | behind>
Fresh-context review check — scan for either a file under .agentic/reviews/* whose name references the PR's commit range or number, or a gh pr view --json reviews entry with state: APPROVED.
Linked task / ADR — scan commit message bodies under <base>..HEAD and the PR body for task-NNNN, ADR-NNNN, spec-NNNN, #<issue>, Closes, Fixes.
Unresolved comments — count entries from gh api that lack a resolved flag or carry an in-progress thread state.
Phase 3 — decision. Apply the bar:
- CI failing → hard stop. Refuse to merge until CI is green, unless the user explicitly overrides ("merge anyway"). On override, log a loud warning that this is a deliberate CI-failing merge and the responsibility is the user's. Per ADR-0025 §3, even the hard gate yields to explicit user authorization, but the override is surfaced visibly.
- CI pending → ask the user: wait for CI, or proceed anyway? Default = wait.
- CI green + warnings (no fresh-context review / no linked task / unresolved comments) → surface each warning, ask the user to confirm the merge, proceed on confirm.
- All green → proceed.
State the decision back to the user before Phase 4 so they can interject.
Phase 4 — merge. Detect repo's allowed merge modes:
gh repo view --json mergeCommitAllowed,squashMergeAllowed,rebaseMergeAllowed
Decision tree:
- Exactly one mode allowed → use it.
- Multiple modes allowed → ask the user: "Repo allows squash / rebase / merge-commit. Pick one." Wait for their choice.
- None allowed (rare) → surface the policy error and stop.
Run the merge:
gh pr merge <num> --squash --delete-branch
(replace --squash with --rebase or --merge per the chosen mode).
--delete-branch by default for feature branches (feat/*, fix/*, chore/*, docs/*, refactor/*). Skip --delete-branch if the source branch is a long-lived integration branch (e.g., cli, develop, release/*).
Capture and report the merge commit URL.
<output_contract> The output is a merged PR. The skill returns:
- The merge commit URL.
- A one-line summary (
<count> commits merged into <base> via <mode>). - The list of warnings the user proceeded past (if any), for the audit trail. </output_contract>
Next
- After merge: pull the latest base locally (
git checkout <base> && git pull). - If the merge surfaced a recurring drift:
ad-driftfor systemic gaps, or updateWORKFLOW.md§10 / §11. - If the merge closed a task: confirm the task file's Status is
doneand the Notes log captures the merge commit URL. - If the merge shipped a binding decision worth recording:
ad-adr(three-criteria rule).
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.
- 4d ago First seen · 103 lines · 134 tokens per session scan A d836a1f4c53b
ad-merge is a skill published in the GitHub repository CorridorTech/PoseCap (190 stars, last pushed 11d ago), licensed Apache-2.0. It adds 134 tokens to every session and 1,306 once invoked, about $0.0007 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.
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…