human-writing
Warn
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill is configured to fetch dynamic writing guidelines from a third-party URL (https://guidelines.sh/api/guidelines?type=writing) every time it is used. This introduces a dependency on external content that is not controlled by the skill itself.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted remote content and applies it as a set of rules for the AI's behavior, creating a surface for indirect prompt injection attacks.
- Ingestion points: The agent fetches data from guidelines.sh which is then used to direct text analysis and rewriting.
- Boundary markers: None; the skill does not instruct the agent to ignore or isolate potentially malicious instructions contained within the fetched guidelines.
- Capability inventory: The skill utilizes WebFetch to retrieve remote data and ReadFile to process local files.
- Sanitization: There is no evidence of validation or sanitization for the content retrieved from the remote API.
Audit Metadata