plan-recovery

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local CLI tool named <pw> for diagnostic and recovery tasks such as <pw> help recovery, doctor, and validation commands.
  • [PROMPT_INJECTION]: The skill processes untrusted external data (project state, manifests, and result artifacts) which could contain embedded instructions, creating a surface for indirect prompt injection.
  • Ingestion points: The skill reads state.json, manifest.json, project-graph.json, and content in the results/ directory as described in SKILL.md.
  • Boundary markers: Absent. The skill does not provide instructions to the agent to treat this data as untrusted or to use delimiters.
  • Capability inventory: The skill allows file system modification and execution of the <pw> CLI tool as specified in SKILL.md.
  • Sanitization: Absent. The skill does not describe any filtering or escaping of the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 10:36 PM
Security Audit — agent-trust-hub — plan-recovery