Keyword: lodestar
- @qmilab/lodestar-shipReleased11h agoVersion0.5.0
Ship a session's raw event log to a remote collector as the versioned lodestar.session_ship@1 NDJSON wire format, with the locked sensitivity ceiling applied client-side before egress. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-cognitive-coreReleased11h agoVersion0.5.0
Claim extraction, evidence linking, and belief adoption for the Lodestar epistemic chain — the layer above the Memory Firewall. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-coreReleased11h agoVersion0.5.0
Schemas and types for the Lodestar epistemic chain. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-event-logReleased11h agoVersion0.5.0
Append-only NDJSON event log writer and reader for the Lodestar epistemic chain. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-action-kernelReleased11h agoVersion0.5.0
Tool registry, two-phase action execution, and sandbox profiles for the Lodestar epistemic chain. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-memory-firewallReleased11h agoVersion0.5.0
Lifecycle gates for claims and beliefs — four orthogonal axes (truth, retrieval, security, freshness) with per-axis transition rules. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-traceReleased11h agoVersion0.5.0
The read side — project the Lodestar event log into the epistemic chain and render a 'why did the agent do this?' report. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-adapter-filesystemReleased11h agoVersion0.5.0
Filesystem adapter for the Lodestar Action Kernel: governed reads and root-confined writes. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-adapter-gitReleased11h agoVersion0.5.0
Git status tool adapter for the Lodestar Action Kernel. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-adapter-shellReleased11h agoVersion0.5.0
Governed shell-command tool adapter for the Lodestar Action Kernel. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-adapter-nostrReleased11h agoVersion0.5.0
Governed Nostr transport tool adapter (publish/fetch) for the Lodestar Action Kernel. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-adapter-httpReleased11h agoVersion0.5.0
Governed HTTP transport tool adapter (fetch/request) for the Lodestar Action Kernel. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-memory-firewall-mem0Released11h agoVersion0.5.0
mem0 import adapter — bring mem0 records through the Lodestar Memory Firewall as external_document evidence at unverified/restricted, never silently promoted. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-memory-firewall-zepReleased11h agoVersion0.5.0
Zep fact-import adapter — bring Zep facts through the Lodestar Memory Firewall as external_document evidence at unverified/restricted, never silently promoted (Zep ratings never raise truth status). Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-memory-firewall-lettaReleased11h agoVersion0.5.0
Letta memory-block import adapter — bring Letta blocks through the Lodestar Memory Firewall as external_document evidence at unverified/restricted, never silently promoted. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-policy-kernelReleased11h agoVersion0.5.0
The Policy Kernel — compiles a declarative, signable action Policy into the Action Kernel's PolicyGate. Three-valued gate (allow/deny/hold), trust-ladder floor, and the approval-request lifecycle. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-harnessReleased11h agoVersion0.5.0
The Lodestar Harness — probe packs, sentinels, and calibrators that exercise and audit the epistemic chain. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-guardReleased11h agoVersion0.5.0
The write-side trust layer — wrap an agent loop and route every tool call through the Action Kernel, every observation through the Cognitive Core. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-guard-mcpReleased11h agoVersion0.5.0
MCP proxy mode for Lodestar Guard — wrap any MCP-speaking agent (Claude Code, Cursor, Aider) so every tool call passes through the Action Kernel and every result through the Cognitive Core. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-runtime-coreReleased11h agoVersion0.5.0
The language-agnostic governance-gate sidecar — govern a non-MCP agent runtime (LangGraph, CrewAI, AutoGen) by remoting each native tool call through the Action Kernel over a thin NDJSON-RPC seam. The reusable spine of Lodestar's runtime-adapter epic.
- @qmilab/lodestar-viewerReleased11h agoVersion0.5.0
The read-side Governing UI — a local, read-only web viewer over the Lodestar event log. Renders the live epistemic chain and trust report in the browser. Part of Lodestar, the trust layer for AI agents.
- @qmilab/lodestar-otel-exporterReleased11h agoVersion0.5.0
Project the Lodestar event log into OpenTelemetry GenAI spans and export them over OTLP/HTTP — pair the epistemic chain with Langfuse, Phoenix, Jaeger, or Tempo. Part of Lodestar, the trust layer for AI agents.