skills/harshanandak/forge/portability/Gen Agent Trust Hub

portability

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Beads issue stores and JSONL snapshots into the Forge kernel environment.
  • Ingestion points: Data is read from local directories via forge migrate --from beads --source <dir> and forge export --import.
  • Boundary markers: The instructions do not specify any delimiters or warnings for the agent to ignore potentially malicious instructions embedded within the imported JSONL data.
  • Capability inventory: The agent is granted access to the Bash tool and terminal to execute the forge CLI, allowing it to perform file operations and data processing.
  • Sanitization: The skill body does not describe any sanitization, validation, or filtering of the external content before it is processed by the kernel.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 10:07 AM
Security Audit — agent-trust-hub — portability