agent-pd: An Audit and Logging Tool for Monitoring Claude Code Agent Behavior
By
softie123
Summary
agent-pd is a logging-only hook and CLI tool for Claude Code agents that functions like a "police department" — it records every tool and permission event from the main agent and its subagents, then replays that log through six detectors to report rule offenses (such as permission bypass, out-of-scope actions, and credential misuse) with quoted evidence. It operates on a catch-and-report basis, never blocking actions, making it a flight recorder and police scanner rather than a firewall.
Source
Key quotes
· 4 pulledA police department for your Claude Code agents
Catch-and-report — it never blocks.
agent-pd won't stop the heist — but every move your agents make ends up on the record.
Flight recorder + police scanner, not a firewall.
You might also wanna read
Agent Bar: Menu Bar Interface for Claude Code AI Projects
Agent Bar is a desktop application that provides a native menu bar interface for Claude Code, allowing users to interact with AI projects th
AgentPeek: Monitor Claude Code and Codex Sessions from Your Mac Notch
AgentPeek is a Mac utility that lets users monitor Claude Code and Codex sessions directly from the Mac notch. It tracks sessions, permissio
Claude Code Launches Multi-Agent AI Code Review System for Bug Detection
Anthropic's Claude Code now offers a multi-agent AI code review system that analyzes pull requests to catch bugs, security issues, and logic
claude-devtools: Open-source tool visualizes hidden Claude Code session data
claude-devtools is an open-source tool that reads raw Claude Code session logs from a user's machine and reconstructs all the information th
Tracea: Self-hosted monitoring and debugging tool for AI agents with automatic RCA and team memory
Tracea is a self-hosted monitoring and debugging tool for AI agents, similar to Datadog but purpose-built for agent workflows. It captures a
Building Custom Coding Agents: A Technical Deepdive into LangChain's Agentic Harness
This article explains that the true power of coding agents like Claude Code comes not from the AI model itself, but from the "agentic harnes
pub.towardsai.net·2d agoBuilding Custom Coding Agents: A Technical Deepdive into LangChain's Agentic Harness
This article explains that the true power of coding agents like Claude Code comes not from the AI model itself, but from the "agentic harnes
pub.towardsai.net·2d agoComments
Sign in to join the conversation.
No comments yet. Be the first.
