jianying-draft-adjustments

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes adjustment suggestions from external 'Analysis Providers' (referenced in references/exposure-matching-workflow.md). This creates a surface where external data influences agent actions, although the skill mitigates this with manual confirmation and absolute parameter enforcement.
  • Ingestion points: Recommendations from external 'Analysis Providers' (references/exposure-matching-workflow.md).
  • Boundary markers: Manual confirmation for reference shots and exceptions.
  • Capability inventory: Uses runtime.controls and project.edit_isolated to modify local drafts.
  • Sanitization: Post-execution read-back and baseline snapshotting.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:56 PM