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.
git clone --depth 1 https://github.com/skyfox675/agents-skillsWrote 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/commands/skyfox675/agents-skills/rescue)<a href="https://agentmods.dev/commands/skyfox675/agents-skills/rescue"><img src="https://agentmods.dev/badge/commands/skyfox675/agents-skills/rescue.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 5.1 | $0.00049 | $0.01164 |
| Opus 5 | $0.00024 | $0.00582 |
| Sonnet 5 | $0.00010 | $0.00233 |
| Haiku 4.5 | $0.00005 | $0.00116 |
Grade A, and why
rescue 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 7d 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/rescue — unstick a PR or an environment
Operate in caveman mode (load the
cavemanskill) to conserve tokens. Keepghcommands, GraphQL, check/job names, and PR numbers byte-exact; compress only the narration. Any issue you file goes through thehumanizerskill (see the issue-filing skill).
Arguments: $ARGUMENTS
Parse the single argument: a number (with or without #) is a PR rescue; anything else (dev, staging, prod, an ephemeral env name) is an environment rescue. No argument: survey owned open PRs and the integration branch for the most blocked thing and propose the target — don't guess silently.
The protocols live in the driving-prs-to-merge skill (PR lifecycle, CI triage ladder, thread etiquette, conflict recovery) and its CI-ownership rules — consult it before acting. Diagnosis precedes mutation: most "stuck" PRs are false signals that cost nothing to clear, and a wrong rescue burns CI runs and review attention.
PR rescue (/rescue <pr#>)
-
Snapshot the PR:
gh pr view <N> --json state,mergeStateStatus,autoMergeRequest,labels,headRefName+gh pr checks <N>+ unresolved review threads via the GraphQL query in driving-prs-to-merge (--json reviewThreadscan return empty — use raw GraphQL). If the PR is already merged or closed, stop and say so — fixes pushed to merged branches are orphaned (re-land via fresh branch if work is needed). -
Rule out false signals first, per the skill's catalog: stale cancelled-run shadow (a newer SUCCESS run for the same SHA), queued-not-broken (
CLEAN+autoMerge=false→ re-rungh pr merge <N> --auto; "already queued" is confirmation), silently-dropped auto-merge (re-arm, it's idempotent), green-but-stuck (almost always unresolved bot threads, not the queue). -
Failing checks: read the failing job's raw log via the API — summary and gate jobs only show the cascade, not the cause. Classify: transient/flake (does the same check pass on sibling PRs and the integration branch?) → one rerun before touching code; real failure → fix in the PR's existing worktree/branch, following the pre-push hygiene from dispatching-subagents (rebase first, reinstall deps, regenerate derived artifacts).
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.
- 7d ago First seen · 42 lines · 49 tokens per session scan A 518c4e71fa65
rescue is a command published in the GitHub repository skyfox675/agents-skills (10 stars, last pushed 7d ago), licensed MIT. It adds 49 tokens to every session and 1,164 once invoked, about $0.0002 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 2026-08-31.
Other commands, from other repositories
composite-actions
Generate, review, secure, and test composite GitHub Actions following best practices — full repo scaffold, interview-driven generation, PR creation on existing repos, SHA pinning, secrets-as-inputs, job summaries, and actionlint validation.
land-and-deploy
Merge PR, wait for CI, verify deploy, run canary — the complete landing pipeline.
devops
You are the DevOps Specialist -- the single authority on version control, CI/CD pipelines, platform operations (GitHub, Azure DevOps, GitLab), infrastructure, deployment, backup, and cleanup. If it touches git, pipelines, or production infrastructure, it's yours.
ci
Run CI validation, create PRs, and merge on green.
ci
Trigger GitHub Actions CI/CD builds, sign, publish, or check CI status.
release
Standalone SDK release command for the BUILD repo. Not a workspace phase — runs independently after any number of implement/redteam cycles. Handles PyPI publishing, documentation deployment, and CI management for the kailash Python SDK and its framework packages.