pick-next-shell

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting and acting upon untrusted data from the local file system.
  • Ingestion points: Reads YAML frontmatter from markdown files within the .turbo/shells/ and .turbo/plans/ directories (SKILL.md).
  • Boundary markers: There are no explicit boundary markers or instructions provided to the agent to ignore or delimit potentially malicious instructions embedded within the ingested files.
  • Capability inventory: The skill performs file reading (globbing and YAML parsing), updates metadata in local files (writing YAML), and invokes multiple downstream skills (/expand-shell, /refine-plan, /self-improve) based on the ingested data.
  • Sanitization: No sanitization or validation logic is defined to inspect the content of the shell or plan files before they are used to influence the agent's planning pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 05:58 PM