writing-stickiness
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted user content and processes it for analysis and revision. \n
- Ingestion points: User-provided draft text is read in SKILL.md (Step 1.1) and resources/success-model.md (Step 1). \n
- Boundary markers: Absent. The skill provides no instructions to use delimiters or ignore commands embedded within user data. \n
- Capability inventory: The skill performs file-write operations (creating writer-stickiness-analysis.md) and file-read operations to execute its logic. \n
- Sanitization: Absent. User input is processed directly without validation or escaping. \n- [COMMAND_EXECUTION]: The skill requires the ability to write and read files on the local filesystem (writer-stickiness-analysis.md). While these operations support the skill's legitimate workflow, they involve active interaction with the host environment's file system.
Audit Metadata