session_start

9 registry entries were observed exposing a tool named session_start — none of them flagged as a privileged capability.

Spelled session_start, session.start in the wild — all one tool for the purposes of this page.

Every row is point-in-time scan output, dated below. Forge has an observed tool surface for 8,470 of 26,658 scannable entries (32%) — an entry that has not been scanned cannot appear here, so absence from this list is not evidence that an entry does not expose this tool.

  1. trust 85/100 · ACollectionobserved · static extraction
    Run Claude Code, Codex, Gemini, Cursor Agent and custom coding CLIs as one unified runtime for claw-style agent systems. Runs standalone, wi
  2. trust 85/100 · AMCP serverobserved · static extraction
    Shared memory & coordination for AI coding agents — a blackboard, not an orchestrator.
  3. trust 60/100 · BMCP serverobserved · static extraction
    Memory Palace MCP server for AI agents
  4. trust 60/100 · BMCP serverobserved · static extraction
    Persistent memory + causal root-cause analysis for AI coding agents. 123 MCP tools. EU, GDPR.
  5. trust 60/100 · BMCP serverobserved · static extraction
    Eyes and hands on real Windows PCs — observe, click, type via Glasswarp API.
  6. trust 60/100 · BMCP serverobserved · static extraction
    Persistent, verified, cross-session memory for AI agents. 26 tools. SQLite + sync.
  7. trust 60/100 · BMCP serverobserved · static extraction
    Persistent institutional memory for AI coding agents. Memory that compounds.
  8. trust 60/100 · BMCP serverobserved · static extraction
    Goal-driven Playwright browser sessions for LLMs, with full action traces and aria-snapshots.
  9. trust 60/100 · BMCP serverobserved · static extraction
    Drive a running React Native or Expo app from inside its JS runtime: see, act, prove.

All indexed toolsSearch the registry for session_start