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/nikicat/secrets-dispatcher/commitnpx skills add nikicat/secrets-dispatcher --skill commitgit clone --depth 1 https://github.com/nikicat/secrets-dispatcherWhat 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 | $0.00016 | $0.00473 |
| Opus 5 | $0.00008 | $0.00236 |
| Sonnet 5 | $0.00003 | $0.00095 |
| Haiku 4.5 | $0.00002 | $0.00047 |
Grade A, and why
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.
What it actually says
Commit, Push, and Verify CI
Step 1: Pre-commit loop
Run make pre-commit and fix any failures. This target runs make check (formatting, linting, static analysis) and make test (Go tests + E2E) in parallel.
- If formatting fails: run
make fmtand re-run - If Go vet/staticcheck/tests fail: fix the code and re-run
- If frontend lint/check fails: fix the code and re-run
- Repeat until
make pre-commitexits 0
Step 2: Commit
- Run
git statusandgit diffto review all changes - Stage all relevant files (prefer explicit file names over
git add -A) - If
$ARGUMENTSis provided, use it as the commit message. Otherwise, draft a message from the changes. - Commit. Include
Co-Authored-By: Claude Opus 4.6 <[email protected]>trailer.
Step 3: Push
Push the current branch to origin:
git push
Step 4: Wait for CI
Poll GitHub checks until they complete:
gh run list --branch $(git branch --show-current) --limit 1 --json status,conclusion,databaseId
- If all checks pass: done. Print the run URL and exit.
- If any check fails: proceed to step 5.
Step 5: Fix CI failures
- Fetch the failed run logs:
gh run view <run-id> --log-failed - Diagnose and fix the failure
- Run
make pre-commitlocally to verify the fix - Amend the commit:
git add <fixed-files> git commit --amend --no-edit - Force push:
git push --force-with-lease - Go back to step 4. Give up after 3 CI fix attempts and ask the user for help.
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 · 60 lines · 16 tokens per session scan A 1b3e14119a5f
commit is a skill published in the GitHub repository nikicat/secrets-dispatcher (11 stars, last pushed 28d ago), licensed MIT. It adds 16 tokens to every session and 473 once invoked, about $0.0001 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
commit-guidelines
Dirty repo, uncommitted changes, commit, amend, split, changeset, PR history: inspect repository state and prepare atomic Git commits or jj changesets. Use whenever the worktree is already dirty or you are about to commit or rewrite history.
github-commit-list
List recent commits on a branch, with author and message.
commit
Workflow Phase 11 — Commit Preparation and Execution. Stages and commits the work. Requires /grant-commit first (Git Commit Guard enforces a 15-min ad-hoc consent window; inside a workflow the grant is slug-scoped and one grant covers the whole landing).
commit-planner
Split a dirty working tree into single-concern Conventional Commits. The deterministic inventory.mjs helper groups dirty paths by concern (docs, source + paired test, config, governance); main context refines the grouping and presents a commit plan. Read-only until the user approves the plan — it never stages, never…
programmatic-agent-runs
Govern Cursor SDK local, cloud, self-hosted, and subagent coding runs before they create branches or PRs.
commit-sequence
Split a large AccessFlow working tree into a sequence of logical, individually-reviewable commits — proposing the groups, confirming each with the user, then staging and committing. Knows which file sets must never be split apart (an engine version bump and its connector pin, a DTO constraint and its form rule).…