Infrastructure for safe, observable
AI development

Open-source tools that make AI coding agents safer, more transparent, and easier to trust in production workflows.

Available

tracine-guard

Production-grade security hooks for Claude Code. Validates bash commands, protects credentials, enforces git safety, scopes subagents.

440+ tests Zero deps Claude Code plugin
/plugin marketplace add TracineHQ/tracine-guard
View on GitHub
Available

tracine-convo

Behavioral analytics CLI for AI coding sessions. Indexes conversation logs into SQLite, queries tool calls, mines anti-patterns, tracks session velocity.

2,890 lines of tests Zero deps PyPI + Plugin
pip install tracine-convo
View on GitHub
Coming Soon

triage

AI-assisted SAST triage. Prioritizes static analysis findings with LLM reasoning, reduces false positive noise, accelerates security review.

LangGraph pipeline Multi-tool SAST PyPI

Why Tracine

Safe by default

Pre-built guardrails that validate every tool call before execution. Block dangerous commands, protect credentials, enforce git discipline.

Observable

Query what your AI agent actually did, not what it said it did. Full-text search across sessions, anti-pattern detection, velocity tracking.

Zero dependencies

Pure Python stdlib. No bloat, no supply chain risk, no transitive vulnerabilities. Install and trust.