prompt-master
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input as its primary function, creating an inherent surface for indirect prompt injection. This risk is minimized by the skill's lack of access to dangerous capabilities or sensitive data.
- Ingestion points: User-provided prompt ideas and drafts in SKILL.md and chat context.
- Boundary markers: The skill generates XML-tagged outputs for specific tools but lacks explicit sanitization of initial user input.
- Capability inventory: No tools or commands for network access, file writing, or command execution are identified.
- Sanitization: Absent.
- [EXTERNAL_DOWNLOADS]: The documentation (README.md) suggests cloning the repository from GitHub. GitHub is a well-known service, and the use of the repository is consistent with the skill's distribution and installation purpose.
- [METADATA_POISONING]: A documentation discrepancy exists between the provided author metadata (alexdcd) and the identity mentioned in the README and LICENSE files (Nidhin Joseph Nelson). This appears to be a minor attribution inconsistency and does not pose a security threat.
Audit Metadata