matt-pr
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data to generate PR content, creating a potential surface for indirect prompt injection.
- Ingestion points: The agent is instructed in
SKILL.mdto read../team-workflow/references/matt-integration.mdand use output from thetw briefcommand to populate the PR template. - Boundary markers: The instructions lack explicit boundary markers or warnings to ignore instructions that might be embedded within the project knowledge or integration rules.
- Capability inventory: The skill is entirely instructional and does not include any custom scripts, file-write operations, or network calls. It relies on the agent's base capabilities to read documentation and execute project-specific CLI tools.
- Sanitization: There is no evidence of sanitization or filtering applied to the external context before it is interpolated into the PR body template.
- [NO_CODE]: The skill is comprised solely of markdown instructions and YAML metadata. No executable scripts, binaries, or automated configuration files are included in the package.
Audit Metadata