ln-222-story-replanner

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill parses existing story content to determine replanning operations, creating a vulnerability if the data contains malicious instructions.\n
  • Ingestion points: Full story descriptions are fetched via get_issue (Linear mode) or Read (file mode) during Phase 1 of the workflow.\n
  • Boundary markers: The workflow does not define specific delimiters or isolation techniques to ensure the agent treats fetched story content strictly as data rather than instructions.\n
  • Capability inventory: The skill possesses significant write capabilities, including save_issue, create_issue, create_comment, Edit, Write, and mkdir, which are executed in Phase 5 based on the comparison logic.\n
  • Sanitization: There is no evidence of sanitization or filtering of ingested story text to prevent embedded prompts from manipulating the operation categorization (KEEP/UPDATE/OBSOLETE/CREATE).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 07:50 AM