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/mitodl/agent-kit/deploy-verificationnpx skills add mitodl/agent-kit --skill deploy-verificationgit clone --depth 1 https://github.com/mitodl/agent-kitWhat 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 | $0.00127 | $0.01710 |
| Opus 5 | $0.00063 | $0.00855 |
| Sonnet 5 | $0.00025 | $0.00342 |
| Haiku 4.5 | $0.00013 | $0.00171 |
Grade A, and why
deploy-verification 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 yesterday.
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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deploy Verification
A config or manifest change merging is not the same as it taking effect. Three separate things can each silently fail while the previous one succeeds: the CD pipeline can be pending or red, the pods can never actually restart even though the pipeline ran, and the restarted pods can still be running the old value if the change didn't reach the path you think it did. This skill walks all three, in order, and demands evidence at each step rather than inferring success from the step before it.
Core rule: report pass/fail per step with the raw evidence (a pod age, a rollout status, a query result, a preview diff) — "looks deployed" or "should be live now" is not a verification, it's a guess with better formatting. If a step can't be checked (no cluster access, no query result yet), say so explicitly rather than skipping it silently.
Step 1 — Confirm the CD pipeline ran and succeeded
Find the deploy workflow for the merged PR's repo and confirm a run completed after the merge commit landed, not just that one exists:
gh run list -R <owner>/<repo> --workflow <deploy-workflow>.yml --limit 5 \
--json databaseId,status,conclusion,headSha,createdAt
Match headSha (or the commit it built from) against the actual merge SHA
— a run against a stale SHA doesn't cover this change. status: in_progress
means wait and re-check, not "probably fine." A conclusion: failure stops
here: the change never reached the cluster, and nothing past this step is
worth checking yet.
Step 2 — Confirm the pods actually rolled
A green pipeline only means the deploy step ran, not that the workload picked it up. Check pod age against the deploy time:
kubectl get pods -n <namespace> -l app=<app> -o wide
kubectl rollout status deployment/<name> -n <namespace> --timeout=5m
Always bound rollout status with --timeout. Its default is --timeout=0s,
which kubectl documents as "zero means never" — an unbounded watch, so a
stalled rollout hangs the verification instead of reporting anything. A
non-zero exit on timeout is itself a finding: the rollout did not complete
within the window, which is exactly the "looks deployed but isn't" case this
checklist exists to catch. Report it as a failed step, not as a flaky command
to retry with a longer wait.
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.
- yesterday First seen · 159 lines · 127 tokens per session scan A d027b706ca6f
deploy-verification is a skill published in the GitHub repository mitodl/agent-kit (2 stars, last pushed 3d ago), licensed BSD-3-Clause. It adds 127 tokens to every session and 1,710 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 2026-08-31.
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…