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/voidnxsec/securellm-mcp/agents-mdgit clone --depth 1 https://github.com/VoidNxSEC/securellm-mcpWhat 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 | $0.04539 | $0.04539 |
| Opus 5 | $0.02269 | $0.02269 |
| Sonnet 5 | $0.00908 | $0.00908 |
| Haiku 4.5 | $0.00454 | $0.00454 |
Grade A, and why
securellm-mcp 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 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 — 1,169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Purpose
This file defines how AI agents should operate inside this repository.
The goal is to produce reliable, auditable, secure, and maintainable changes with minimal noise and maximum respect for the existing architecture.
Agents should act like careful engineering collaborators: understand the system first, make scoped changes, verify their work, and leave a clear handoff.
1. Core Operating Principles
1.1 Respect the Existing System
Before changing code, understand:
- What the current architecture is trying to preserve.
- Which conventions are already present.
- Which modules own which responsibilities.
- Which abstractions are stable and which are still experimental.
- Which files are generated, vendored, or externally managed.
Prefer extending existing patterns over introducing isolated new ones.
1.2 Optimize for Coherence
Every change should improve or preserve:
- Architectural coherence.
- Security posture.
- Operational reliability.
- Testability.
- Reproducibility.
- Observability.
- Developer experience.
A technically clever solution that breaks the shape of the system is usually worse than a smaller coherent change.
1.3 Work in Small, Auditable Steps
Prefer small, reviewable patches.
Each meaningful change should have:
- A clear reason.
- A clear scope.
- A clear verification path.
- A clear rollback path when possible.
1.4 Treat Instructions as a Hierarchy
When instructions conflict, follow this priority order:
- Explicit user request for the current task.
- Security, privacy, and safety requirements.
- This
AGENTS.md. - More specific nested
AGENTS.mdfiles. - Existing project conventions.
- General best practices.
Nested AGENTS.md files apply to their directory subtree.
2. Agentic Workflow
Use this loop for every non-trivial task:
INTAKE
↓
UNDERSTAND CONTEXT
↓
MAP CURRENT STATE
↓
PLAN
↓
IMPLEMENT
↓
VERIFY
↓
DOCUMENT / HANDOFF
2.1 Intake
Clarify internally:
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 · 1,169 lines · 4,539 tokens per session scan A ee998a311cd1
securellm-mcp AGENTS.md is an instructions file published in the GitHub repository VoidNxSEC/securellm-mcp (0 stars, last pushed 27d ago), licensed Apache-2.0. It adds 4,539 tokens to every session, about $0.0227 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 instructions, from other repositories
minio-mcp-server CLAUDE.md
Claude Code instructions for saeedshoh/minio-mcp-server, covering claude.md, what this is, setup & running, configuration and architecture.
adeu GEMINI.md
Instructions for dealfluence/adeu, covering adeu — native track changes for ai, available tools, readdocx, processdocumentbatch and applytextrevision.
gradle-mcp AGENTS.md
Instructions for rnett/gradle-mcp, covering gradle mcp project memory, primacy zone: fundamental mandates, technical standards (openspec), build & test commands and misc tips.
openings-mcp AGENTS.md
Instructions for amikai/openings-mcp, a project described as: 💼 One MCP server to search job boards and company career sites.
raindrop-mcp dxt.instructions.md
Instructions for adeze/raindrop-mcp: I want to build this as an MCP Bundle, abbreviated as "MCPB". Please follow these steps.
mcpscore AGENTS.md
Instructions for mcp-box/mcpscore, covering agent instructions for mcpscore and gotchas (learned the hard way).