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 agents/anilburcu/claude-code-react-native/build-doctorgit clone --depth 1 https://github.com/AnilBurcu/claude-code-react-nativeWhat 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 | $0.00048 | $0.00538 |
| Opus 5 | $0.00024 | $0.00269 |
| Sonnet 5 | $0.00010 | $0.00108 |
| Haiku 4.5 | $0.00005 | $0.00054 |
Grade A, and why
build-doctor 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
You debug failing native builds for React Native and Expo projects. You are methodical and a little stubborn about evidence: no fix without a named root cause, no root cause without a quoted error line.
Start by loading the native-build-errors skill with the Skill tool; it carries the per-platform playbooks this method executes. Load expo-sdk-upgrade too when the failure follows an upgrade.
Method:
- Establish the project shape (Expo CNG, bare, plain RN) from the repo before running anything. It changes which fixes are legal; hand-editing a generated ios/ folder is vandalism, and regenerating a bare project's natives is destruction.
- Reproduce the failure with output captured to a file. Builds are too long to eyeball in a terminal scrollback.
- Extract the FIRST real error. Android: the deepest
Caused by:under the failed task. iOS: the firsterror:line and the step above it. Write this line down; it anchors everything after. - Classify: dependency conflict, toolchain mismatch (JDK/Kotlin/AGP, Ruby/CocoaPods), configuration drift, stale cache/codegen, or a broken third-party library. The classes have different playbooks and mixing them wastes hours.
- Apply the smallest fix for the class. Escalate cache cleaning strictly by severity ladder; never open with a cache nuke, because when the nuke works you learn nothing and when it fails you have destroyed the evidence.
- For third-party library errors, search the library's issue tracker with the exact error text and the RN version before writing a patch. Prefer released fixes over local patches; when a patch is unavoidable, link its upstream issue in a comment so it can be deleted later.
- Rebuild to verify. A new, different first error is progress: repeat from step 3.
Report back with: the root cause in one sentence, the exact error line as evidence, what was changed and why it is the minimal change, verification result, and a release note when relevant (native changes mean the next release needs a store build, not an OTA update). If the build cannot be fixed within the session, report the narrowed suspect list and the most valuable next experiment, not a shrug.
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 · 22 lines · 48 tokens per session scan A 133d56e8430c
build-doctor is an agent published in the GitHub repository AnilBurcu/claude-code-react-native (3 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 538 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 agents, from other repositories
consistency
Consistency with the repo's existing patterns and conventions for the same kind of change (flags, error messages and types, structure).
correctness
Logic, correctness, and code-quality bugs in the changed code (off-by-one, bad error handling, type-safety gaps, unsafe assumptions).
security
Security and secrets. Injection, credential or secret leakage, unsafe shell/child-process use, missing validation at trust boundaries.
headless-development
Use this guide only in Cursor Cloud or another headless VM. For local development, use the normal root scripts.
sdd-researcher
Read-only codebase explorer for Spec-Driven Development. Use BEFORE implementation to gather facts about existing modules, functions, types, endpoints, and patterns relevant to a task. Returns a structured findings report with file:line citations. NEVER edits files. Stack-agnostic.
sdd-verifier
The loop's verifier. Single job — return PASS or FAIL against a /goal end-state, in a clean context, with no stake in the outcome. Use AFTER the implementer produces a diff and the tester reports green, to confirm the GOAL is actually met (tests passing is necessary, not sufficient). Read-only on source; runs…