Live context engine for AI agents. Open source infrastructure for verifiable, audit-ready agent state.
Perseus — Live Context Engine
Renders workspace state into static facts before your AI agent's context window opens. No orientation tax. No stale docs. Just instant, verifiable context. MIT licensed. On-premise.
Mimir — Persistent Memory
Zero-dependency persistent memory for AI agents. SQLite + FTS5. AES-256-GCM encrypted. 43 MCP tools. Sub-millisecond recall. Single portable mimir.db file.
Plutus — Credit Monitoring
Multi-provider LLM credit balance tracking. Real-time balance checks, burn-rate forecasts. Never run out of credits mid-session.
MCTS — Security Scanner
Model Context Threat Scanner. Static analysis for MCP servers. 31 analyzers catching injection, exposed secrets, unverified tool inputs. Run before you trust an MCP server.
PR Pilot — Autonomous Review
5-agent pipeline: reviewer → fixer → tester → verifier → escalator. Autonomous PR review that finds bugs, writes fixes, verifies deployments, and escalates when blocked.
Blast Radius — Impact Analysis
GitLab Orbit-powered dependency graph analyzer. See exactly what breaks before you change code.
- SBOMs published per EO 14028
- NIST AI RMF aligned (all four functions)
- EAR99 self-classified — no ITAR restrictions
- Air-gap ready — zero cloud dependencies
- MIT licensed — no vendor lock-in