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 commands/generous-corp/pulp/doctorgit clone --depth 1 https://github.com/Generous-Corp/pulpWrote 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/generous-corp/pulp/doctor)<a href="https://agentmods.dev/commands/generous-corp/pulp/doctor"><img src="https://agentmods.dev/badge/commands/generous-corp/pulp/doctor.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 | $0.00008 | $0.00791 |
| Opus 5 | $0.00004 | $0.00396 |
| Sonnet 5 | $0.00002 | $0.00158 |
| Haiku 4.5 | $0.00001 | $0.00079 |
Grade A, and why
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 today.
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
Check the development environment for missing dependencies and configuration issues.
Prepend the output with an Environment section that includes the canonical version line (see .claude/commands/version.md for the parsing recipe):
Environment
Claude plugin <plugin_version> · Pulp SDK/CLI <sdk_version>
Then run the full doctor:
pulp doctor
Options:
pulp doctor --fix— auto-fix issues where possiblepulp doctor --ci— CI mode, exit codes onlypulp doctor --dry-run— show what--fixwould dopulp doctor android/pulp doctor ios— run mobile development environment checkspulp doctor --versions [--scan-parents] [--json]— diagnose CLI / SDK / plugin version skewpulp doctor --validators [--fix] [--dry-run]— verify auval / pluginval / clap-validator install and code-signature healthpulp doctor --caches [--fix] [--dry-run] [--json]— audit and heal the FetchContent shared-source cachepulp doctor gpu [--no-render] [--json]— perform bounded render/readback and compute/map work;--no-renderis an unverified inventory/preflight that acquires no GPU devicepulp doctor --host-quirks/pulp doctor quirks— show the runtime DAW host-quirks policy and enforced accommodationspulp doctor --au-cache --dry-run— preview macOS AudioComponentRegistrar refresh after AU metadata changespulp doctor list/pulp doctor --only "<name>"— enumerate checks or run one targeted probepulp doctor --only WidgetBridge— check generated WidgetBridge.d.ts, mock function lists, and JS bridge docs for source-tree input-fingerprint freshnesspulp doctor --only "Control broker"— run only the observational local carrier probe
Default checks include C++20 compiler availability, CMake, git/git-lfs, LFS-backed Skia binaries when present, generated WidgetBridge API artifacts in source-tree mode, VST3/AudioUnit SDKs where relevant, optional AAX setup, package/platform alignment, Cmajor when used, build configuration, pulp-mcp availability, and optional local control-broker reachability. Broker absence is nonfatal. reachable-unverified means only that a carrier accepted a connection; without a trusted peer expectation, do not describe it as healthy or verified. This probe must not create or repair the runtime directory, start a daemon, open a session, or acquire authority.
Run this first when builds fail unexpectedly or on a new machine. Run pulp doctor --validators if pulp validate aborts with "broken code signature", pulp doctor --caches if build/test reports FetchContent cache drift, and pulp doctor --host-quirks when DAW accommodations or host-specific runtime behavior look suspicious.
For GPU problems, prefer pulp doctor gpu --json before inspecting a build or
opening a window. Exit 0 means required real-work proofs passed and at least one
required probe has authentic identity; exit 1 is a completed measured failure,
and exit 2 is unavailable or unverified evidence. Report the adapter identity
exactly as Dawn returns it; never infer hardware identity from the backend name
or describe a null/software adapter as a hardware pass. Probe identities are
independent and do not assert same-device correlation. The installed command
and its pulp_gpu_doctor MCP peer work outside a source checkout.
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.
- today First seen · 48 lines · 8 tokens per session scan A 6d89cddb5d74
doctor is a command published in the GitHub repository Generous-Corp/pulp (16 stars, last pushed today), licensed MIT. It adds 8 tokens to every session and 791 once invoked, about $0.0000 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 2026-09-04.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.