writing-for-pstack
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, potentially untrusted sources to generate instructions for subsequent agent runs.
- Ingestion points: The skill reads
AGENTS.mdat the repository root, other repository documents, and operator-provided text drafts. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its instructions and potentially malicious commands embedded within the analyzed files.
- Capability inventory: The output of this skill is a command (
/patina-mode) that invokes an autonomous agent mode, creating a path for injected instructions to influence high-autonomy actions. - Sanitization: The skill lacks logic to sanitize or validate the content extracted from the repository files before incorporating it into the final prompt.
- [COMMAND_EXECUTION]: The primary purpose of the skill is to generate a tool invocation command (
/patina-mode <sentence>) for the agent to execute, which facilitates the transition to an autonomous execution state.
Audit Metadata