sync-workspaces

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external local sources.
  • Ingestion points: The skill reads directory structures and project metadata from configured filesystem 'roots' and a 'registry' file (SKILL.md, Steps 2 and 3).
  • Boundary markers: The skill employs clear logical boundaries, specifically an 'Iron Law' (REPORT ONLY, NO AUTO-CREATE) and a mandatory requirement for human 'grants' or confirmation before any material state changes (ROLE.md; SKILL.md, Step 5).
  • Capability inventory: The agent has the capability to read and scan the local filesystem and write to project registries or configuration files when granted permission.
  • Sanitization: The instructions do not define specific sanitization or filtering procedures for the content or paths read from the filesystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 07:02 AM
Security Audit — agent-trust-hub — sync-workspaces