Weir

Weir is an API key leak detection tool that finds credentials in agent traces and coding sessions.

Weir is an API key leak detection tool designed for AI agents and coding sessions. It tests agent frameworks by planting fake keys and checking spans, logs, console output, and models for exact value matches. For coding agents like Claude Code, it provides a status line to monitor credentials and an optional hook to strip known keys from tool results. The tool runs offline and deterministically without sending data anywhere. It supports Linux and macOS platforms.

Weir at a glance

Best for: Software engineer
Runs on: Linux, macOS, CLI
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: github.com/idogol24/weir

Key features

Plant fake keys to test agent flows
Display a status line for coding agent sessions
Redact keys from tool results with an optional hook
Scan sensitive data through existing traces
List credentials an agent could reach via inventory
Audit real credentials against saved agent sessions

Use it for

Test agent frameworks for API key leaks in CI or locally
Monitor real credentials entering Claude Code sessions
Strip known keys from tool results before models read them
Check real credentials against saved agent sessions offline

Alternatives to Weir

Claude Tap — Inspect prompts, tool calls, and token usage locally without a hosted dashboard.
Claude Scrub — Removes exposed API keys, passwords, and tokens from local AI session history.
Stroq — Scans recorded sessions and guards future actions with local hooks.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
LeakLock — It scans live sites to find keys missing from git history and generates AI-builder fix prompts.
Stop That Shit — Enforces task boundaries to stop unnecessary AI agent actions.
Flash Agents — Offloads reading and typing to cheap workers while keeping Claude for review
Agent Archive — Hand off coding sessions between Claude Code, Codex, and Cursor across machines.
Coucou — It provides open-source, real-time agent monitoring and approvals directly from the screen notch.
Raindrop Workshop — Provides real-time observability and self-healing evaluation loops for local AI agents.
Squeeze — Cuts token usage for coding agents by shrinking verbose tool output while retaining key data.
Ctxfw — Reduces AI context token mass while preserving code structure and dependencies.

All Weir alternatives