skills/marcellocurto/skills/to-spec/Gen Agent Trust Hub

to-spec

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository and conversation history, creating a surface for instructions embedded in those sources to influence the agent.
  • Ingestion points: Processes codebase content and conversation context as defined in SKILL.md.
  • Boundary markers: Lacks clear delimiters or instructions to ignore potential commands within the repository data.
  • Capability inventory: Includes the ability to explore the filesystem and interact with external issue trackers (e.g., via GitHub CLI).
  • Sanitization: No specified mechanisms for filtering or validating the repository content before synthesis.
  • [COMMAND_EXECUTION]: The skill implicitly uses system commands for repository discovery and issue submission.
  • Evidence: Instructions in SKILL.md to 'Explore the repo', 'resolve the publication target... from the Git remote', and 'publish it to the project issue tracker'.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 09:18 AM
Security Audit — agent-trust-hub — to-spec