The deterministic, verifiable grounding layer for AI code work. Flags fabricated files, imports and symbols in AI answers against your repo and installed libraries. Zero dependencies, offline, byte-stable; also a ranked signature map and MCP server.
Allowed — no malicious behavior detected
21 low-signal patterns reviewed and cleared.
AI security review
scanned · by lpm-firewall-ai
The scanner’s lifecycle concern is not supported: the manifest’s hook is publish-only, and the inspected configuration writes are tied to explicit setup commands. No concrete credential transfer, command-output exfiltration, or unconsented install-time agent mutation was established.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Running the explicit --setup or skills install command.
Impact
The inspected behavior supports intended SigMap setup; no confirmed attack surface was established.
Mechanism
These commands write SigMap-managed project or agent configuration for its MCP server, hooks, or skills.
Decision evidence
public snapshot
Evidence for AI clean decision
The manifest points to the core API and CLI binaries and defines a publish-only hook, not an install hook.
Configuration changes are activated by the explicit --setup command; skill installation also requires the explicit skills install command.
The adapter and integration files contain writes for SigMap-managed context, MCP, hook, and skill configuration.
The inspected dashboard, judge, MCP, retrieval, and session modules expose product functions for reporting, grounding, context handling, ranking, and reports.
The project-map entrypoint describes generating a project map.
Automated scan signals; the review checked them and cleared this version. Where they are, most severe file first. Code opens a file's flagged lines, with their evidence, below.
Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration.
gen-context.js
'Add the `permissions.allow` array above to `.claude/settings.json` to activate.',