draft-project-charter

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data.
  • Ingestion points: The procedure in Step 1 directs the agent to read existing documentation, specifically proposals, emails, and briefs (SKILL.md).
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings to disregard instructions potentially embedded in the source documents.
  • Capability inventory: The skill is granted access to Read, Write, Edit, Bash, Grep, and Glob tools.
  • Sanitization: Absent. The agent is directed to use content from external files to populate the charter template without filtering or validation steps.
  • [COMMAND_EXECUTION]: The skill requests the Bash tool in its allowed-tools metadata. While the documented procedure only describes file creation and editing, the inclusion of a shell environment provides a vector for arbitrary command execution that is not required for the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 10:51 PM