cts-hook

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill does not contain any evidence of credential exposure, data exfiltration, or malicious external downloads. All content is localized within the provided markdown file.
  • [PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data (user-provided keywords, paragraphs, or full articles) to generate output. This creates a surface for indirect prompt injection.
  • Ingestion points: User input provided during the interaction as described in Step 1 (Type A, B, and C) of SKILL.md.
  • Boundary markers: The instructions implement a structured extraction process and a confirmation step when processing full articles, which helps bound the agent's behavior, although it lacks explicit XML-style or unique string delimiters for the user content.
  • Capability inventory: The skill has no access to tools or capabilities that would allow it to execute system commands, access the network, or write to the file system.
  • Sanitization: The skill includes an explicit instruction to stop and wait for user confirmation (Step 1, point 6) if the input is a complex article, allowing for human-in-the-loop validation of the extracted themes before hook generation proceeds.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 07:26 AM
Security Audit — agent-trust-hub — cts-hook