story-outliner
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill ingests untrusted story text for analysis, creating a potential surface for instructions embedded in stories to influence the agent.
- Ingestion points: Story text input via the 'Read' tool in SKILL.md.
- Boundary markers: Absent; instructions do not explicitly tell the agent to ignore commands within the source text.
- Capability inventory: Text summarization and outline generation.
- Sanitization: None specified in the instructions or logic.
- [Prompt Injection] (SAFE): No direct injection attempts or behavior overrides were found in the skill's instructions.
- [Remote Code Execution] (SAFE): No use of package managers, remote script downloads, or dynamic execution functions (eval/exec).
- [Data Exfiltration] (SAFE): No network requests to non-whitelisted domains or access to sensitive local file paths (e.g., .ssh, .env) were detected.
Audit Metadata