skills/mgratzer/forge/forge-ship/Gen Agent Trust Hub

forge-ship

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from sources like issue descriptions, URLs, or plan files via the $ARGUMENTS variable. This data is then used to drive code implementation and review processes, creating a surface for embedded instructions to influence the agent's behavior.\n
  • Ingestion points: The $ARGUMENTS parameter which can contain content from Issues or free-text descriptions.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in this skill definition.\n
  • Capability inventory: The skill orchestrates file system modifications (via implementation) and repository operations (commits/issue creation).\n
  • Sanitization: No sanitization or validation of the input content is described before it is processed by the underlying tools.\n- [COMMAND_EXECUTION]: The skill performs actions that involve modifying files, running tests, and managing Git operations by delegating tasks to related skills like forge-implement. While standard for a developer tool, these capabilities can be abused if the input source is compromised.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 01:26 PM
Security Audit — agent-trust-hub — forge-ship