Agentic-AI-Cyber-Security: Instructions file for Claude Code

CLAUDE.md

Agentic-AI-Cyber-Security CLAUDE.md is an instructions file for Claude Code from Krishcalin/Agentic-AI-Cyber-Security. It costs 6,452 tokens per session, scanned B, original, MIT.

Project instructions for an open-source Python security analyzer that checks source code for security flaws, fake dependencies, and prompt-injection attacks, with AI-based code analysis.

In plain words
What is it for?
Use them when developing or reviewing this analyzer, including its scanning engines, security rules, MCP integrations, command-line tools, configuration, and CI/CD workflows.
Why use it?
They give a coding agent the project's architecture, design principles, supported components, and development context so changes fit the existing security system.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions Claude Code.

This is Krishcalin/Agentic-AI-Cyber-Security's own configuration. It tells Claude Code how to work on Agentic-AI-Cyber-Security itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Agentic-AI-Cyber-Security configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Krishcalin/Agentic-AI-Cyber-Security. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Krishcalin/Agentic-AI-Cyber-Security/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Krishcalin/Agentic-AI-Cyber-Security

Made for: Claude Code.

Wrote 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.

agentmods badge for Agentic-AI-Cyber-Security CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/krishcalin/agentic-ai-cyber-security/claude-md.svg)](https://agentmods.dev/instructions/krishcalin/agentic-ai-cyber-security/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/krishcalin/agentic-ai-cyber-security/claude-md"><img src="https://agentmods.dev/badge/instructions/krishcalin/agentic-ai-cyber-security/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 6,452 This file is loaded in full into every session.
When invoked 6,452 The same file — it is already loaded in full.
Security scan B 3 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.06452 $0.06452
Opus 5 $0.03226 $0.03226
Sonnet 5 $0.01290 $0.01290
Haiku 4.5 $0.00645 $0.00645

Measured 7d ago against content hash c2ee471e7b46, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade B, and why

Agentic-AI-Cyber-Security CLAUDE.md scanned grade B with 3 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 7d 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.

Instruction-override phrasingmediumPrompt injection

Text telling the model to disregard its earlier instructions or safety rules is the shape of a prompt injection, whoever wrote it.

python main.py scan-prompt --text "Ignore previous instructions and..."

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- [x] Dockerfile rules (16) — :latest, root, secrets, curl|bash

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- Sinks: `os.system()`, `subprocess.run()`, `cursor.execute()`, `eval()`
CLAUDE.md · 519 lines

How it starts

The opening of the file, as written. The whole thing — 519 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — Agentic AI Cyber Security

Project Overview

An open-source Python-based source code security analyzer that identifies security flaws, flags fictitious or non-existent dependencies, prevents prompt injection attacks, and delivers AI-driven semantic code analysis — accessible through MCP integrations with Claude Code or command-line interfaces and CI/CD pipelines.

Repository: https://github.com/Krishcalin/Agentic-AI-Cyber-Security License: MIT Python: 3.10+ Status: All phases + P1/P2/P3 complete — MITRE ATLAS + OWASP LLM Top 10 mapped Rules: 441 across 19 languages | Engines: 24 | MCP Tools: 30 | CLI Commands: 25+ Inspired by: sinewaveai/agent-security-scanner-mcp


Architecture

Directory Structure

Agentic-AI-Cyber-Security/
├── config/                            # Configuration files
│   ├── settings.yaml                  # Global scanner settings
│   └── profiles/                      # Scan profiles (quick, full, ci)
│       ├── quick.yaml
│       ├── full.yaml
│       └── ci.yaml
├── core/                              # Core engine components
│   ├── __init__.py
│   ├── engine.py                      # Main scanner orchestrator
│   ├── ast_analyzer.py                # AST-based vulnerability detection (Python)
│   ├── pattern_matcher.py             # Regex/pattern-based scanning (multi-lang)
│   ├── taint_tracker.py               # Cross-function taint flow analysis
│   ├── package_checker.py             # Dependency verification (PyPI, npm, crates)
│   ├── prompt_scanner.py              # Prompt injection detection engine
│   ├── semantic_reviewer.py           # LLM-powered code review (Claude API)
│   ├── fix_generator.py               # Auto-fix template engine
│   ├── mcp_auditor.py                 # MCP server security auditor (Tier 1)
│   ├── rag_scanner.py                 # RAG pipeline security scanner (Tier 1)
│   ├── tool_response_analyzer.py      # Tool response injection analyzer (Tier 1)
│   ├── chain_detector.py              # Multi-step exploit chain detector (Tier 2)
│   ├── policy_engine.py               # Declarative YAML policy engine (Tier 2)
│   ├── runtime_monitor.py             # Real-time session anomaly detection (Tier 2)
│   ├── redteam_generator.py           # Adversarial test suite generator (Tier 2)
│   ├── dependency_analyzer.py         # Supply chain dependency analyzer (Tier 2)
│   ├── reporter.py                    # Report generation (terminal, JSON, SARIF, HTML)
│   ├── grader.py                      # A–F security grading system
│   ├── models.py                      # Data models (Finding, ScanResult, Severity)
│   └── logger.py                      # Structured logging
├── rules/                             # YAML security rules (organized by language)
│   ├── python.yaml                    # Python-specific rules
│   ├── javascript.yaml                # JavaScript/TypeScript rules
│   ├── java.yaml                      # Java rules
│   ├── go.yaml                        # Go rules
│   ├── php.yaml                       # PHP rules
│   ├── ruby.yaml                      # Ruby rules
│   ├── c_cpp.yaml                     # C/C++ rules
│   ├── dockerfile.yaml                # Dockerfile rules
│   ├── terraform.yaml                 # Terraform/IaC rules
│   ├── kubernetes.yaml                # Kubernetes manifest rules
│   ├── typescript.yaml                # TypeScript rules (34) — Tier 2
│   ├── shell.yaml                     # Shell/Bash rules (30) — Tier 2
│   ├── rust.yaml                      # Rust rules (30) — Tier 2
│   ├── swift.yaml                     # Swift/iOS rules (30) — Tier 2
│   ├── kotlin.yaml                    # Kotlin/Android rules (30) — Tier 2
│   ├── prompt_injection.yaml          # Prompt injection patterns
│   └── common.yaml                    # Cross-language rules (secrets, hardcoded creds)
├── mcp_server/                        # MCP (Model Context Protocol) server
│   ├── __init__.py
│   ├── server.py                      # MCP server entry point (stdio transport)
│   ├── tools.py                       # MCP tool definitions and handlers
│   └── schemas.py                     # Input/output JSON schemas for tools
├── cli/                               # CLI interface
│   ├── __init__.py
│   └── main.py                        # Click-based CLI entry point
├── integrations/                      # CI/CD and editor integrations
│   ├── github_actions.py              # GitHub Actions reporter
│   ├── gitlab_ci.py                   # GitLab CI integration
│   └── sarif_exporter.py              # SARIF 2.1.0 export for Code Scanning
├── data/                              # Static data files
│   ├── pypi_packages.bloom            # Bloom filter — PyPI package names
│   ├── npm_packages.bloom             # Bloom filter — npm package names
│   ├── crates_packages.bloom          # Bloom filter — crates.io package names
│   └── known_malicious.yaml           # Known malicious package list
├── templates/                         # Report templates
│   ├── report.html                    # HTML report template (Jinja2)
│   └── fix_templates/                 # Auto-fix templates by CWE
│       ├── cwe_78.py                  # OS Command Injection fixes
│       ├── cwe_89.py                  # SQL Injection fixes
│       ├── cwe_79.py                  # XSS fixes
│       ├── cwe_798.py                 # Hardcoded Credentials fixes
│       └── ...
├── tests/                             # pytest test suite
│   ├── conftest.py
│   ├── test_engine.py
│   ├── test_ast_analyzer.py
│   ├── test_pattern_matcher.py
│   ├── test_taint_tracker.py
│   ├── test_package_checker.py
│   ├── test_prompt_scanner.py
│   ├── test_semantic_reviewer.py
│   ├── test_fix_generator.py
│   ├── test_grader.py
│   ├── test_mcp_server.py
│   ├── test_rules/                    # Rule validation tests
│   └── fixtures/                      # Vulnerable code samples per language
│       ├── python_vulnerable.py
│       ├── javascript_vulnerable.js
│       └── ...
├── benchmarks/                        # Performance and accuracy benchmarks
│   ├── accuracy_test.py               # Precision/recall against known CVEs
│   └── results.md                     # Benchmark results
├── main.py                            # CLI entry point
├── pyproject.toml                     # Project metadata + dependencies
├── requirements.txt                   # Pinned dependencies
├── CLAUDE.md                          # This file
└── README.md

Read the full file on GitHub · 519 lines

Changes

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.

  1. 7d ago First seen · 519 lines · 6,452 tokens per session scan B c2ee471e7b46

Subscribe to this mod's changes

Agentic-AI-Cyber-Security CLAUDE.md is an instructions file published in the GitHub repository Krishcalin/Agentic-AI-Cyber-Security (1 stars, last pushed 5mo ago), licensed MIT. It adds 6,452 tokens to every session, about $0.0323 per session on Opus 5. A static security scan graded it B with 3 findings (instruction-override phrasing, makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

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.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

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).

microsoft/vscode · 6,785 tokens

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).

microsoft/vscode · 5,001 tokens

langchain AGENTS.md

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.

langchain-ai/langchain · 4,469 tokens

spec-kit AGENTS.md

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.

github/spec-kit · 7,104 tokens