doctor

A troubleshooting command for Audacity, a desktop application for recording and editing audio, and its coding-agent connection. It checks the plugin, server, communication pipes, and Audacity, then explains likely fixes.

In plain words
What is it for?
Use it before investigating failed audio operations, timeouts, empty results, incorrect exports, or a connection that will not start.
Why use it?
It gives a structured answer when an audio tool fails, hangs, returns nothing, or produces an incorrect export. It also identifies missing Python-related setup and connection problems.

Command

Install

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.

agentmods
npx agentmods add commands/robworks-code/audacity-mcp-max/doctor
Clone the repo
git clone --depth 1 https://github.com/robworks-code/audacity-mcp-max
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,249 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00034 $0.01249
Opus 5 $0.00017 $0.00624
Sonnet 5 $0.00007 $0.00250
Haiku 4.5 $0.00003 $0.00125

Measured 2d ago against content hash b34bc3a3d670, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 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.

commands/doctor.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/audacity:doctor

Answer "why isn't this working" across both halves of the plugin. Run this before anything else when a tool call fails, times out, returns an empty result, or an export comes out wrong.

Step 1: the plugin side

The MCP server cannot report on the environment that launched it, so check that first:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/plugin_doctor.py"

On an old-python3 machine this re-runs itself under uv run, and a first run there builds the venv before it can print anything, so it can take a minute or two; after that it is instant. If it produces no output at all, or the shell reports python3: command not found or opens a Command Line Tools install prompt, this machine has no python3 - which is a supported state, since uv is what provides Python here. Tell the user to install uv from https://docs.astral.sh/uv/ and re-run, or run the script with the plugin's own interpreter ("${CLAUDE_PLUGIN_ROOT}/.venv/bin/python") if the server has ever started.

Read the output and note anything that is not in a good state:

  • uv: not found - the server never started. This is the whole problem. Tell the user to install uv from https://docs.astral.sh/uv/ (or set UV_BIN to its path) and restart their MCP client. Stop here; nothing below will work.
  • venv built: no - normal before the first successful launch. Harmless on its own, but combined with a working uv it means the server has not started yet this session. Note: on an old-python3 machine, plugin_doctor.py itself re-runs under uv run, and uv run builds the venv as a side effect of that re-exec - so running this diagnostic can be what makes venv built flip to yes, not just launching the server. That's expected, not a sign of anything wrong.
  • mod-script-pipe: disabled / ask / absent - Audacity will not create the pipes. Point the user at /audacity:setup.
  • mod-script-pipe: no-config - no audacity.cfg was found at all, meaning Audacity has never been launched on this machine (or its config was reset). Tell the user to launch Audacity once so it writes its config, then run /audacity:setup.
  • pipe (to) or pipe (from) missing - Audacity is not running, or was not restarted after the module was enabled.
  • python: ... is too old - python3 on this machine is older than 3.10 (stock macOS still ships 3.9), and there was no plugin venv and no uv to re-run under. The install is fine; the interpreter is not. Tell the user to install uv from https://docs.astral.sh/uv/ - that also fixes uv: not found above, and the server needs it anyway. Everything below this line in the report will say unknown or unavailable, so treat none of it as a finding.
  • pipe and config info: unavailable - the report could not get far enough to check the pipes, and nothing below it is trustworthy. Two causes, and the detail in the parentheses says which:
    • (see the python line above) - the old-interpreter case above. Fix that, not the install.
    • anything else - the plugin's own files are broken. Reinstall the plugin.
  • measurement: not installed - pipelines will still run, but their reports will show lufs: null and any loudness target as unknown rather than met or missed. Not a fault; it is an optional extra. Point the user at /audacity:setup --measurement if they want loudness verification.
  • measurement: unknown - the check could not be run, which on this script means the venv interpreter could not be probed. Treat it exactly like the transcription unknown case above.

Read the full file on GitHub · 94 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. 2d ago First seen · 94 lines · 34 tokens per session scan A b34bc3a3d670

Subscribe to this mod's changes

doctor is a command published in the GitHub repository robworks-code/audacity-mcp-max (0 stars, last pushed 18d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,249 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.