Borrowing it
Nothing to install: this file belongs to adamayoung/TMDb. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/adamayoung/TMDb/main/.claude/skills/fix-integration-failures/SKILL.mdgit clone --depth 1 https://github.com/adamayoung/TMDbWrote 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/adamayoung/tmdb/fix-integration-failures)<a href="https://agentmods.dev/skills/adamayoung/tmdb/fix-integration-failures"><img src="https://agentmods.dev/badge/skills/adamayoung/tmdb/fix-integration-failures/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/adamayoung/tmdb/fix-integration-failures"><img src="https://agentmods.dev/badge/skills/adamayoung/tmdb/fix-integration-failures.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00069 | $0.02870 |
| Opus 5.5 | $0.00028 | $0.01148 |
| Sonnet 5.5 | $0.00014 | $0.00574 |
| Haiku 4.5 | $0.00007 | $0.00287 |
Grade A, and why
fix-integration-failures 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 1mo 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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fix Integration workflow failures
The Integration workflow (.github/workflows/integration.yml) runs the
live-API integration suite on a weekly schedule (cron: '0 0 * * 0' — Sunday
00:00 UTC), as well as on PRs/pushes. The scheduled run is a live-API canary:
nothing in the code changed, so a failure means the live TMDb API drifted, a test's
assumed data went stale, or a transient error/rate-limit hit. This skill takes such
a failure from red run → green main: diagnose it, then either re-run a transient
or fix the real cause on its own branch and open (and optionally merge) a PR.
Scope. This is for an Integration failure not attached to an open feature PR — the scheduled canary, a
workflow_dispatch, or a failure onmain. When a failing Integration check belongs to an open PR,/watch-prand/fix-pr-checksown that; they delegate the pre-existing/unrelated case back here (fix on a branch offmain), which is exactly this skill's job.
Mode — check the arguments passed to this skill (shown at the end). If they
include merge (e.g. /fix-integration-failures merge), auto-merge the fix PR
once green. Otherwise stop at ready-to-merge and hand off (default).
Agent behaviour contract
- Never edit
maindirectly (CLAUDE.md). Any fix lands on afix/<slug>branch offmain, via a PR. - Diagnose before fixing. Always run
/diagnose-integration-failurefirst; let its ranked cause drive the fix. Don't guess. Check the diagnosis carries its evidence. A shape-change (cause 1) or drifted-data (cause 2) conclusion must arrive with anobserved:line — the live call that saw today's response. Missing one, re-run the diagnosis once asking for it. Still missing → treat that cause as unverified: proceed, but say so in the PR body and inclaude-analysis.md, and never describe it as confirmed drift. Re-run once, not in a loop — a headless run has no MCP and legitimately cannot observe (it reportsobserved: unavailable (headless)), so an unbounded retry would spin forever. Transient (cause 3) and in-diff regression (cause 0) carry noobserved:line by design; do not demand one. - Distinguish transient from real. A re-run is the cheap test. Only open a PR for a cause that survives a re-run (or is clearly a data/shape drift).
- Test-first for real fixes. A model/decoder fix follows
canon-tdd(failing unit test + fixture, then the fix). A drifted-assertion fix updates the integration test to assert behaviour, not a brittle exact value. - The gate is
make ci. Never open the PR until it passes locally. - Don't paper over a real regression. If the cause is a genuine library bug (not drift/transient), fix it properly or stop and report — never just relax a test to hide it.
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.
- 1mo ago First seen · 198 lines · 69 tokens per session scan A b03c7c67bf0b
fix-integration-failures is a skill published in the GitHub repository adamayoung/TMDb (179 stars, last pushed yesterday), licensed Apache-2.0. It adds 69 tokens to every session and 2,870 once invoked, about $0.0003 per session on Opus 5.5. 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
reproduce-bug
Reproduce a reported bug in googleapis/mcp-toolbox and decide whether it is real, delivering an evidence-backed verdict: confirmed, already fixed, misconfiguration, client-side, works as intended, not reproducible, or blocked. Use whenever a maintainer asks you to reproduce, verify, confirm, or investigate a bug…
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
testing-blocks
Use this when you have made AEM Edge Delivery Services code changes to blocks, scripts, or styles and need to validate them before opening a pull request. Covers unit testing for utilities and logic, browser testing with Playwright, linting, and guidance on what to test and how.
reality-verification
This skill should be used when the user asks to "verify a fix", "reproduce failure", "diagnose issue", "check BEFORE/AFTER state", "VF task", "reality check", "check test quality", "mock-only tests", or needs guidance on verifying fixes by reproducing failures before and after implementation, or detecting mock-heavy…
webapp-qa
Perform scoped exploratory QA of a web application, reproduce defects, collect screenshots and console evidence, and deliver a prioritized report.
test-electron-app
Drive the real running PostHog Electron app (live tRPC, workspace-server, real data) over CDP with agent-browser. Connect to the running app on port 9222, test desktop changes against a local Django stack, snapshot the accessibility tree, inspect network requests, and screenshot only when explicitly asked. Use when…