gsd-add-phase

Pass

Audited by Gen Agent Trust Hub on Apr 3, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted user input from {{GSD_ARGS}} to perform privileged file system actions.\n
  • Ingestion points: User-supplied phase descriptions enter the context via the {{GSD_ARGS}} variable in SKILL.md.\n
  • Boundary markers: The skill lacks explicit delimiters or instructions for the agent to ignore potential instructions embedded within the user-provided phase description.\n
  • Capability inventory: The skill has the capability to create new directories, modify existing files (roadmap, STATE.md), and spawn additional agents using the spawn_agent tool.\n
  • Sanitization: While the process describes generating a slug from the description for directory naming, it does not specify sanitization or validation of the input content before it is written to roadmap files or passed to sub-agents.\n- [COMMAND_EXECUTION]: The skill performs automated file system operations including directory creation and file updates.\n
  • Evidence: The objective and process sections explicitly outline that the workflow handles directory creation, roadmap entry insertion, and updates to the STATE.md tracking file.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 3, 2026, 04:33 AM