DhanushNehru

DhanushNehru/agentcheck

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

9 good first / help-wanted issues · JavaScript · last activity Oct 17, 2026

1 stars 0 forks 1 watchers JavaScript MIT License
ai-agents claude-code cli cursor devtools llm-security mcp prompt-injection security
View on GitHub
9 Open Issues Need Help Last updated: Oct 17, 2026

Open Issues Need Help

View All on GitHub
Add a --staged mode to scan only git-staged files 14 days ago
enhancement help wanted
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Add a --diff mode to scan only files changed relative to a git ref (for PR scans in CI) 15 days ago
enhancement good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Discussion: detecting instruction-file drift (when CLAUDE.md no longer matches what the agent does) 16 days ago
enhancement help wanted
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Track a lockfile-style hash of remote MCP server tool lists to catch silent changes 16 days ago
enhancement help wanted
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Test coverage for json-parse-error, binary files and oversized files 18 days ago
good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Add a CONTRIBUTING.md with a "how to add a detection rule" guide 18 days ago
documentation good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
base64url payloads evade the base64-payload rule 18 days ago
enhancement good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Exfil endpoint list misses Discord webhooks and current ngrok domains 18 days ago
enhancement good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security
Add a way to exclude paths from the scan (ignore file or --exclude flag) 18 days ago
enhancement good first issue
DhanushNehru/agentcheck
1

Scan AI coding-agent config files (CLAUDE.md, AGENTS.md, .cursor/rules, MCP configs) for hidden prompt-injection attacks.

JavaScript
#ai-agents#claude-code#cli#cursor#devtools#llm-security#mcp#prompt-injection#security