forge-verify
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
forge-memoryCLI tool for logging results (log), managing data state (consolidate), and synchronization (sync). - [COMMAND_EXECUTION]: The workflow involves running the project's full test suite, which entails executing code found within the environment's test directories.
- [PROMPT_INJECTION]: The skill represents an indirect prompt injection surface as it ingests untrusted data from story files, source code, and external architecture decisions.
- Ingestion points: Processes content from
.forge/sprint-status.yaml, story files,tests/, andsrc/directories. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the processed data are provided.
- Capability inventory: Access to
forge-memoryCLI, file-writing capabilities (writing missing tests), and execution of the test suite. - Sanitization: No evidence of sanitization or validation of the ingested content before it is processed by the agent.
Audit Metadata