start-here

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from the repository to build context.
  • Ingestion points: Project README, documentation introductions, package manifests (package.json, pyproject.toml, go.mod), git commit history (last 15 commits), PR titles, and Open Issues.
  • Boundary markers: Absent. The skill lacks specific instructions or delimiters to prevent the agent from executing instructions potentially hidden within repository documentation or issue tracker content.
  • Capability inventory: The skill requires read access to local files and git metadata, and write access to create the docs/gtm-cofounder/founder-brief.md file.
  • Sanitization: Absent. No sanitization or filtering of the ingested repository content is performed before the agent processes it.
  • [COMMAND_EXECUTION]: The skill instructs the agent to retrieve repository metadata including commit history, PR titles, and open issues. This involves executing shell-based commands (such as git log) or interacting with developer tools to gather project status information.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 01:13 PM
Security Audit — agent-trust-hub — start-here