writing-for-pstack
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from repository-level files to generate commands for another system.
- Ingestion points: The skill reads
AGENTS.mdand other repository documents located at the root of the project to extract work definitions and facts. - Boundary markers: The skill does not define explicit delimiters or 'ignore' instructions for the content read from repository documents.
- Capability inventory: The skill's output is a formatted command (
/patina-mode <sentence>) intended to be executed by a downstream agent. It does not perform network operations or direct file writes itself. - Sanitization: The instructions require the agent to 'Extract facts' and 'Remove instructions' already owned by playbooks, which provides a level of filtering but does not fully sanitize the input against adversarial instructions hidden in prose.
Audit Metadata