forge-quick-spec
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several
forge-memoryCLI commands includingsearch,log,consolidate, andsync. These are local operations designed to manage project state and history within the FORGE development environment. - [DATA_EXFILTRATION]: The skill reads project-specific context from
.forge/memory/MEMORY.md. This is a local file access within the project scope. No network requests to non-whitelisted domains or access to sensitive system credentials (e.g., SSH keys, AWS configs) were detected. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external project data.
- Ingestion points: Data is ingested from
.forge/memory/MEMORY.mdand the user-provided change description. - Boundary markers: No delimiters or instructions to ignore embedded commands are used when reading project memory.
- Capability inventory: Capabilities are limited to executing the specific
forge-memorytool suite. - Sanitization: There is no explicit sanitization of the memory file content or the user's description before they are used in the workflow or interpolated into CLI logging commands.
Audit Metadata