writing-gate-checker

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads project-sensitive files including briefs, drafts, and review logs from the local workspace. However, it does not use any network tools or external URLs, meaning there is no risk of data exfiltration.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from various external project files which could theoretically contain malicious instructions.
  • Ingestion points: Files like 02-Drafts/草稿_v*.md, 03-Reviews/Claim-Ledger.md, and .claude/skills/writing-clone-profile/references/gate-check-contract.md.
  • Boundary markers: Absent; the skill does not implement specific delimiters to isolate processed content from its own instructions.
  • Capability inventory: Limited to reading local markdown files and writing a report to 03-Reviews/Gate-Check-Report.md. It cannot execute shell commands or access the internet.
  • Sanitization: No sanitization or escaping of the ingested file content is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 01:26 AM
Security Audit — agent-trust-hub — writing-gate-checker