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/codegoddy/codeguardian_mcp/claude-mdgit clone --depth 1 https://github.com/codegoddy/codeguardian_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 | $0.01904 | $0.01904 |
| Opus 5 | $0.00952 | $0.00952 |
| Sonnet 5 | $0.00381 | $0.00381 |
| Haiku 4.5 | $0.00190 | $0.00190 |
Grade A, and why
codeguardian_mcp CLAUDE.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 yesterday.
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 — 247 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CodeGuardian MCP - AI Instructions
You have access to CodeGuardian, an advanced code validation tool that catches hallucinations in AI-generated code using AST parsing and prompt engineering best practices.
CRITICAL RULE
ALWAYS call validate_code AFTER generating any code, BEFORE presenting it to the user.
This is non-negotiable. AI-generated code frequently references functions, classes, or methods that don't exist in the user's project. These "hallucinations" compile fine but crash at runtime.
Enhanced Workflow (Chain-of-Thought)
1. User asks for code
2. You generate code
3. THINK: "I need to validate this against the actual codebase"
4. Call validate_code({ projectPath: ".", newCode: "<your code>", language: "typescript" })
5. ANALYZE: Review the validation results
- Score < 70? Critical issues exist
- Hallucinations found? Must fix before presenting
- Suggestions provided? Use them to correct the code
6. If issues found → Fix them using the suggestions → Validate again
7. Present validated code to user with confidence
Validation Confidence Levels
When you receive validation results, interpret them as follows:
- Score 90-100: ✅ High confidence - Code is safe to use
- Score 70-89: ⚠️ Medium confidence - Minor issues, review suggestions
- Score 50-69: 🔴 Low confidence - Multiple issues, fix before using
- Score < 50: ❌ Critical - Do not use, major hallucinations detected
Available Prompts (Use These for Better Results)
CodeGuardian provides multiple prompt templates based on prompt engineering best practices:
1. validate - Basic validation (Zero-shot)
Quick check for hallucinations. Use for simple code snippets.
2. validate-detailed - Chain-of-thought reasoning
Step-by-step validation with detailed reasoning. Use for complex code.
3. validate-with-examples - Few-shot learning
Validation with examples of common AI mistakes. Use when learning patterns.
4. validate-comprehensive - Multi-perspective analysis
Validates from multiple angles (symbols, dependencies, logic). Use for critical code.
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.
- yesterday First seen · 247 lines · 1,904 tokens per session scan A 11b572492004
codeguardian_mcp CLAUDE.md is an instructions file published in the GitHub repository codegoddy/codeguardian_mcp (1 stars, last pushed 6mo ago), licensed MIT. It adds 1,904 tokens to every session, about $0.0095 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
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.