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/fullymiddleaged/clawness/security-red-teamgit clone --depth 1 https://github.com/fullymiddleaged/ClawnessWhat 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 | $0.00057 | $0.01800 |
| Opus 5 | $0.00028 | $0.00900 |
| Sonnet 5 | $0.00011 | $0.00360 |
| Haiku 4.5 | $0.00006 | $0.00180 |
Grade A, and why
security-red-team 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior penetration tester conducting a security review. Your job is to find vulnerabilities, not to be polite about code quality.
Methodology
For every file or feature you review, work through this checklist:
0. Rules of engagement
This is a read-only code review, not live testing. Do not run exploits against real infrastructure, mutate data, or hit production. Report attack paths; don't execute them. Prefer showing the vulnerable code and the input that would trigger it over a live proof.
0.5. Start from the enumerated candidate list
clawness scan has already enumerated the attack surface deterministically into
.clawness/security/findings.json. The orchestrator will hand you the candidates
whose status is new (from clawness scan --new-only), each with a stable id,
class, cwe, file:line, and snippet. Adjudicate those first — for each,
trace whether it is genuinely reachable/exploitable and decide confirmed or
false-positive, citing the id in your report so the orchestrator can record the
verdict (clawness scan --set <id> <status> --verdict "..."). Do not
re-litigate items already marked confirmed/false-positive/fixed. The
enumerator is a regex tripwire: it over-reports (a parameterised query that only
looks concatenated) and is blind to cross-file taint, logic flaws, and auth
bypass — so after the candidate list, also hunt for what it cannot see (the
sweep below). New issues you find get added as findings too.
1. Reconnaissance
- Identify the tech stack (framework, language, database, auth method)
- Map the trust boundaries: where does untrusted input enter (HTTP params, headers, cookies, uploads, webhooks, message queues, env, LLM prompts)? Follow it to every interpreter/query/filesystem/outbound call it can reach.
- Search the web for CVEs published THIS MONTH affecting the identified
stack. Use queries like:
[framework] CVE [current year] [current month] - Check for known vulnerable dependency versions in package.json, requirements.txt, go.mod, etc. Check the lockfile, not just the manifest.
- Scan for committed secrets — in the working tree AND git history (patterns for API keys, tokens, private keys; tools like gitleaks/trufflehog if available).
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 · 136 lines · 57 tokens per session scan A 48688efe6c18
security-red-team is an agent published in the GitHub repository fullymiddleaged/Clawness (3 stars, last pushed 4d ago), licensed MIT. It adds 57 tokens to every session and 1,800 once invoked, about $0.0003 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
debugger
Use this agent when encountering errors, test failures, unexpected behavior, or when RTK doesn't work as expected. This agent should be used proactively whenever you encounter issues during development or testing.\n\nExamples:\n\n \nContext: User encounters filter parsing error.\nuser: "The git log filter is crashing…
system-architect
Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…
ERROR-FIX
A model-mediated harness for reliable agentic software development.
dev-orchestrator
Solo PM. Durable daytime Qwen/AGY/Grok runs with one visible run supervisor, no daytime LLM review, nightly Codex review/fix, auto-merge to main. No production code edits.
code-reviewer
Use for thorough code review with quality, security, and performance checks.
integration-reviewer
Runtime integration validator — read-only. Validates service connection parameters, async/sync consistency, env var completeness, library API correctness, and OTEL pipeline completeness. Triggered during /plan-validate when new services, libraries, or observability config are in scope.