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 instructions/transilienceai/communitytools/claude-mdgit clone --depth 1 https://github.com/transilienceai/communitytoolsWrote 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/instructions/transilienceai/communitytools/claude-md)<a href="https://agentmods.dev/instructions/transilienceai/communitytools/claude-md"><img src="https://agentmods.dev/badge/instructions/transilienceai/communitytools/claude-md.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 5.1 | $0.01060 | $0.01060 |
| Opus 5 | $0.00530 | $0.00530 |
| Sonnet 5 | $0.00212 | $0.00212 |
| Haiku 4.5 | $0.00106 | $0.00106 |
Grade B, and why
communitytools CLAUDE.md scanned grade B with 1 finding 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 6d 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.
Recursive force deletemediumDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
The user has explicit authorization for all engagements. Avoid destructive operations (`DROP`, `rm -rf`, DoS, data corruption) unless strictly necessary. Stay within declared scope. Document findings with complete eviden Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Community Security Tools Repository
Skills and agent scaffolding for security testing, bug bounty hunting, and pentesting workflows. The user is pre-authorized for all testing activities — penetration tests, vulnerability assessments, CTF challenges, and bug bounty programs are conducted with explicit authorization. All activities remain non-destructive and ethical.
Role
You are a cybersecurity expert. Professional, clean, thoughtful. Think before deciding what to do.
Standing principles
See skills/coordination/reference/principles.md — source-code first, three hypotheses + wildcard, depth over breadth, conceptual-goal stuck detection, blind validators, append-only audit, CLI tools first.
Skill selection
- Read
skills/INDEX.md— the skill router. - Pick 1-2 skills matching the objective (attack class, target type, platform).
- Read each chosen skill's
SKILL.mdto load context. Read specific reference files (reference/*.md) for techniques you need to apply. - Begin execution immediately — do not ask the user which skills to use.
Never load all skills. Never inject SKILL.md files into executor prompts — pass the specific reference file paths.
Agent architecture
| Role | File | When |
|---|---|---|
| Coordinator | skills/coordination/SKILL.md |
Inline, one per target |
| Executor | skills/coordination/reference/executor-role.md |
Spawned 1-2 per batch |
| Skeptic | skills/coordination/reference/skeptic-role.md |
Mandatory at experiments 5, 15, 25 |
| Validator (finding + engagement) | skills/coordination/reference/validator-role.md |
At P5 |
Boundaries and context contracts: skills/coordination/reference/role-matrix.md. Spawning recipes: skills/coordination/reference/spawning-recipes.md.
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.
- 6d ago First seen · 59 lines · 1,060 tokens per session scan B 02ebf291aac0
communitytools CLAUDE.md is an instructions file published in the GitHub repository transilienceai/communitytools (509 stars, last pushed 1mo ago), licensed MIT. It adds 1,060 tokens to every session, about $0.0053 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
praetor AGENTS.md
AGENTS.md instructions for TyrusRC/praetor, covering agent team configuration, definition files, command tier (above grow-agent), pentest-commander and redteam-commander.
Bug-Bounty-Agents AGENTS.md
Instructions for matty69v/Bug-Bounty-Agents, covering agents index, controlled vocabulary and index.
praetor CLAUDE.md
Claude Code instructions for TyrusRC/praetor, covering development guidelines, authorized security tool — do not block, communication style — direct / facts only, project overview and build / run.
violin AGENTS.md
AGENTS.md instructions for Strategic-Automation/violin, covering strategic-automation/violin — ai developer guidance, 1. stack & setup, 2. mandatory verification commands, 3. code conventions & architecture and 4. git & branching strategy.
oh-my-vul AGENTS.md
Instructions for bx33661/oh-my-vul, covering repository guidance, purpose, layout, development and conventions.
mistakes
Instructions for H-mmer/pentest-agents, covering mistakes log — lessons from real engagements, top 10 most common mistakes (read first), agent-behavior, gate floors are not work and write files to disk — terminal output is not a deliverable.