gtm-persona

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied criteria and sources to generate persona definitions in markdown format. These files are stored in the workspace and are intended to be read by automated workflows, creating a surface where user-supplied content could potentially influence downstream processes.
  • Ingestion points: User criteria and user-provided sources used to define persona fields (SKILL.md).
  • Boundary markers: The agent follows a specific structure based on templates/PERSONA.md.
  • Capability inventory: Writing to the personas/ directory and Git operations (pull, commit, push) (SKILL.md).
  • Sanitization: The skill does not explicitly specify sanitization or validation of the user-provided criteria beyond adhering to the markdown template.
  • [COMMAND_EXECUTION]: The skill uses Git commands (pull, commit, push) to manage the version history and persistence of persona files within the repository.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for the user to install a dependent skill set (gtm-workspace) using an npx command targeting the author's repository (eliasstravik/gtm-skills).
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 01:11 PM