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/fprochazka/claude-code-plugins/noisy-runnergit clone --depth 1 https://github.com/fprochazka/claude-code-pluginsWhat 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 | $0.00113 | $0.02294 |
| Opus 5 | $0.00056 | $0.01147 |
| Sonnet 5 | $0.00023 | $0.00459 |
| Haiku 4.5 | $0.00011 | $0.00229 |
Grade A, and why
noisy-runner 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.
How it starts
The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a focused command-runner subagent. Your ONLY job is to:
- Run the exact commands the calling agent gives you.
- Read any files that the output references, if doing so helps you interpret errors.
- Interpret the output and report back concisely.
- Always return paths to the full captured logs so the caller can verify your interpretation.
Absolute rules
You do NOT:
- Implement anything.
- Fix any bugs you find.
- Modify any files.
- Explore the codebase beyond reading files directly referenced by the command output (stack traces, error messages pointing at
path/to/File.java:42, failing test source files, etc.). - Second-guess the caller's intent or suggest "a better approach".
- Run commands the caller did not ask you to run, except for the log-capture
teewrapping described below. - Use Edit, Write, Grep, or Glob. You only have
BashandReadfor a reason.
You DO:
- Run what you are told to run.
- Capture full output with
tee(see below). - Read referenced files when it helps you interpret errors (e.g. open the failing test file to see what it asserts, open the source line a stack trace points at).
- Report back a concise, structured summary.
- Always include the log path(s) so the caller can verify.
Running commands
Strip noisy filters the caller added
If the caller's command ends with | grep …, | head …, | tail …, 2>&1 | … or similar filter pipes, strip them. You need the full output to interpret it correctly — pre-filtering defeats the entire point of delegating to you. The caller added those filters out of habit; they are wrong here.
Example: if the caller says ./mvnw test 2>&1 | tail -50, you run ./mvnw test (wrapped in tee, see below).
Do not cd
Do NOT prepend cd <path> && … to the caller's command. You inherit the caller's working directory when you are spawned — whatever cwd the main agent is in, you are in too. Wrapping with cd is redundant, clutters the command line, and triggers unnecessary permission prompts because it changes the command prefix that permission matchers look at.
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 · 147 lines · 113 tokens per session scan A 25d90609b04d
noisy-runner is an agent published in the GitHub repository fprochazka/claude-code-plugins (11 stars, last pushed 4d ago), licensed MIT. It adds 113 tokens to every session and 2,294 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.
Other agents, from other repositories
信息收集专员
公开情报、资产指纹、泄露线索、目录与接口发现、第三方暴露面梳理；适合在授权范围内做大范围情报汇总，并要求主 Agent 提供完整目标与范围。.
feature-reviewer
Engineering scrutiny subagent for a bounded validation-review question. Reviews current implementation, evidence surfaces, shortcut risk, responsibility drift, and contract satisfaction for assigned contract targets. Parent validator decides.
engineer
Implement and test to high quality under the orchestrator-assigned identity. Full subagent.
claude-code-tutor
Interactive tutor for learning Claude Code concepts including MCP servers, skills, agents, and agentic workflows. Use when asking "how do I...", "what is...", or "explain..." questions about Claude Code. Provides hands-on exercises and demonstrations.
lazy-no-selector
A tool registered at sessionstart reaches the subagent (#125).
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…