TMDb: Agent for Claude Code

.claude/agents/check-diagnoser.md

check-diagnoser is an agent for Claude Code from adamayoung/TMDb. It costs 125 tokens per session (891 once invoked), scanned A, original, Apache-2.0.

A diagnostic helper for one failing pull-request status check in the TMDb repository. It identifies the likely cause and suggested fix for either a continuous-integration check or a live-API integration check, without changing the code.

In plain words
What is it for?
Use it to investigate one failed lint, build, test, markdown, or integration check and receive only a concise summary, likely cause, and suggested fix.
Why use it?
It keeps diagnosis separate from fixing, so the cause can be reviewed before anyone applies a change.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions subagents; names the NotebookEdit tool.

This is adamayoung/TMDb's own configuration. It tells Claude Code how to work on TMDb itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything TMDb configures →

Reuse

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.

Copy the file
curl -O https://raw.githubusercontent.com/adamayoung/TMDb/main/.claude/agents/check-diagnoser.md
Clone the repo
git clone --depth 1 https://github.com/adamayoung/TMDb

Made for: Claude Code.

Wrote 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.

agentmods badge for check-diagnoser

README.md
[![agentmods](https://agentmods.dev/badge/agents/adamayoung/tmdb/check-diagnoser.svg)](https://agentmods.dev/agents/adamayoung/tmdb/check-diagnoser)
Your own site
<a href="https://agentmods.dev/agents/adamayoung/tmdb/check-diagnoser"><img src="https://agentmods.dev/badge/agents/adamayoung/tmdb/check-diagnoser.svg" alt="Measured on agentmods" height="20"></a>
Per session 125 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 891 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00125 $0.00891
Opus 5 $0.00063 $0.00445
Sonnet 5 $0.00025 $0.00178
Haiku 4.5 $0.00013 $0.00089

Measured 9d ago against content hash d36cfb9c0145, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

check-diagnoser 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 9d 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.

.claude/agents/check-diagnoser.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Subagent: Check Diagnoser (diagnose, never fix)

You diagnose exactly one failing PR status check for the TMDb repo and report the cause and fix concisely. You are the diagnosis step of /fix-pr-checks; the act-on-it layer — apply → verify → commit → push — is the caller's, never yours.

Your task names the check, the branch, and the routed skill:

  • /diagnose-ci-failure — any CI check: lint, markdown lint, build, or unit tests from ci.yml.
  • /diagnose-integration-failure — the Integration check (live-API suite from integration.yml).

Use that skill to diagnose: it locates the failing run, reads the log, and maps it to a cause and fix. On a repeat, the task also carries the prior attempt's Cause/Fix and why it didn't stick — do not re-propose it; find what that diagnosis missed.

The Haiku pin is a floor — the caller escalates

model: haiku above is the first-attempt default, not a ceiling: a call-site model: beats the frontmatter pin (ADR-0014), so /fix-pr-checks escalates a repeat to Opus by passing model: opus with the spawn, per its run ledger. The pin exists for the spawn that forgets model — it lands on cheap Haiku instead of inheriting the session model.

DO NOT BUILD OR RUN TESTS

No make, no swift build, no swift test, and do not invoke /build, /test or /integration-test. Diagnose by reading the failing run's log and the source. The routed skill tells you to reproduce locally; that step is the caller's, not yours — it owns the single build slot for this worktree. Reproduction here would collide with it and with any sibling diagnosis running beside you.

Half of this is enforced, half is not — know which. The frontmatter's disallowedTools removes Edit/Write/NotebookEdit (you report a diagnosis, you never apply the fix) and Agent (no fanning out — note this also unplugs the build skills' tooling-runner, but those skills then fall back to direct make, so the removal blocks fan-out, not builds). Skill and Bash you keep — invoking the routed diagnosis skill is the job, and its log reads (gh, jq) need the shell — so make, swift build and the build/test skills are still physically reachable, and the rule above rests on you.

Read the full file on GitHub · 65 lines

Changes

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.

  1. 9d ago First seen · 65 lines · 125 tokens per session scan A d36cfb9c0145

Subscribe to this mod's changes

check-diagnoser is an agent published in the GitHub repository adamayoung/TMDb (177 stars, last pushed 8d ago), licensed Apache-2.0. It adds 125 tokens to every session and 891 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 2026-08-30.

Related

Other agents, from other repositories

retrospective

Reflective analyst who extracts learnings through structured retrospective frameworks, diagnosing agent performance, identifying error patterns, and documenting success strategies. Uses Five Whys, timeline analysis, and learning matrices. Use when you need root-cause analysis, atomicity scoring, or to transform…

rjmurillo/ai-agents · 62 tokens

ci-debugger

CI pipeline debugging specialist. Analyzes CI logs, identifies failure causes, and provides specific fixes for GitHub Actions and GitLab CI.

dykyi-roman/awesome-claude-code · 32 tokens

ci-doctor

Diagnoses and repairs an eligible CI failure on the existing pull-request branch.

theam/facility · 19 tokens

debug

Debug your application to find and fix a bug. Systematic root cause analysis through structured phases: assessment, investigation, resolution, and quality assurance.

rjmurillo/ai-agents · 31 tokens

factory-ci-doctor

Diagnostician for one red GitHub Actions run. Spawn it after a run has failed — never to wait for one (gh run watch --exit-status does that for free) — with the repo and run ID or PR number, so the failed-job logs never enter the caller's context. It returns the culprit job/step, the offending log lines, and a TICKET…

watt-mind/factory · 108 tokens

orc-ci-investigator

Investigator role — diagnoses a failing CI run end-to-end and returns a classified failure report with a concrete fix list; never edits code, never re-runs workflows. Fetches run/job logs via gh CLI, classifies each failure as test / lint / build / flake / infra / environment-drift, and emits fix items orc-code-fixer…

HigorAlves/orc · 118 tokens