outliner

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill ingests untrusted external data (user articles/tweets) which is then used to drive further agent actions and sub-tasks. Ingestion points: External content is received and saved to 'source-1.md' (SKILL.md, Step 0). Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the user material. Capability inventory: The skill performs file system writes and invokes external sub-agents ('writer-agent') using the 'Task' tool (SKILL.md, Step 4). Sanitization: Absent; the content is passed directly to analysis and writing agents without validation.
  • File System Operations (SAFE): The skill creates directories and saves markdown files to 'posts/YYYY/MM/DD/[slug]/'. This is a standard capability required for the skill's stated purpose of content management and does not show signs of malicious path traversal or sensitive file access.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 03:24 PM