ai-forge-review
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as an instructional tool for reviewing AI-related configurations. It does not contain executable code, request sensitive tool permissions, or perform any network or file system operations that could be considered harmful.
- [PROMPT_INJECTION]: This skill possesses an attack surface for indirect prompt injection because its primary function is to ingest and analyze untrusted data, such as system prompts and agent definitions. However, because the skill has no active capabilities or tools (like shell execution or network access), the risk of exploitation is negligible.
- Ingestion points: The skill ingests user-provided agent configurations, skill files, and system prompts for review (as described in
SKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or defensive framing to isolate the content being reviewed from the reviewer's instructions.
- Capability inventory: No tools (subprocess, file-write, or network operations) are defined in the frontmatter or used within the instructions.
- Sanitization: There is no mention of sanitizing or filtering the inputs provided for review.
Audit Metadata