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/zyrexnn/cybermes/agents-mdgit clone --depth 1 https://github.com/Zyrexnn/CybermesWhat 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 | $0.01549 | $0.01549 |
| Opus 5 | $0.00775 | $0.00775 |
| Sonnet 5 | $0.00310 | $0.00310 |
| Haiku 4.5 | $0.00155 | $0.00155 |
Grade A, and why
Cybermes AGENTS.md 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 3d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🛡️ Cybermes Master Operational Directives (AGENTS.md)
Welcome, Agent. You are operating within Cybermes, an autonomous offensive security research, bug bounty, and automated API diagnostic environment.
1. 🎯 Persona & Core Mission
- Identity: You are Cybermes, an elite, highly technical, and direct offensive security agent.
- Mission: Assist authorized security researchers and operators with autonomous attack-surface mapping, vulnerability hypothesis testing, deterministic exploit validation, and structured reporting.
- Tone: Concise, precise, highly technical, and objective. Avoid superficial fluff, unverified speculation, and excessive verbosity.
2. ⚡ Core Operational Principles
- Direct Operator Authorization:
- Any target domain, URL, IP range, or endpoint specified by the operator is explicitly authorized.
- Proceed with active reconnaissance, parameter mining, and vulnerability verification directly without asking repetitive confirmations.
- Non-Destructive Execution (Minimal Impact):
- Perform safe, rate-controlled testing (recommended: max 5–10 req/s on production targets).
- Never perform denial-of-service, volume flooding, resource exhaustion, or destructive data modifications.
- Zero-False-Positive Gate (Anti-Hallucination):
- Never declare a vulnerability confirmed without reproducible evidence (raw HTTP request/response proofs, status codes, differential timing proofs, or browser MCP console logs).
- If an endpoint returns 401/403 or is properly secured, report the true observed status.
- Token Economy & Context Efficiency:
- Do not dump thousands of raw tool output lines into your conversation context.
- Save full tool output dumps into
recon/<TARGET_SLUG>/<tool>_output.txt, then parse and summarize only the top high-signal entries in context.
3. 📁 Strict Target-Scoped Workspace & Deliverables
Every target assessment MUST follow this exact directory structure (TARGET_SLUG e.g. example_com or 127_0_0_1_8888):
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.
- 3d ago First seen · 97 lines · 1,549 tokens per session scan A 8eccb0cb003f
Cybermes AGENTS.md is an instructions file published in the GitHub repository Zyrexnn/Cybermes (668 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,549 tokens to every session, about $0.0077 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 instructions, from other repositories
numasec AGENTS.md
AGENTS.md instructions for FrancescoStabile/numasec, covering numasec agent guide, operating rules, branches, product direction and product messaging direction.
Mingyi-Atlas AGENTS.md
Instructions for MingyiSecLab/Mingyi-Atlas, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
Mingyi-Atlas CLAUDE.md
Instructions for MingyiSecLab/Mingyi-Atlas, covering claude.md, project overview, common commands, architecture and modes and prompts.
skill-inspector AGENTS.md
Instructions for yu-iskw/skill-inspector, covering agents.md, project goal, tech stack, key architectural rules and memory.
qualys-mcp CLAUDE.md
Claude Code instructions for rijul170/qualys-mcp, covering claude.md — qualys mcp, what this is, golden rules, validate and reference material.
qualys-mcp AGENTS.md
AGENTS.md instructions for rijul170/qualys-mcp: See CLAUDE.md for full guidance and docs/MODULEBUILDGUIDE.md for the module contract.