project-links
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides instructions for a documentation workflow to organize project resources and links. It does not contain executable code, remote downloads, or credential exposure patterns.- [NO_CODE]: No scripts, binaries, or executable code patterns were detected in the provided skill content.- [INDIRECT_PROMPT_INJECTION]: The skill involves reading and appending data to Markdown files (e.g.,
20_项目/<项目名>/外部链接.md), creating a surface for processing potentially untrusted data. However, the risk is minimal as the skill lacks high-privilege capabilities such as shell execution or network requests based on that content. - Ingestion points: Reads and appends to files located at
20_项目/<项目名>/外部链接.md. - Boundary markers: Absent.
- Capability inventory: Limited to file appending/writing through agent tools; no subprocess, network, or eval capabilities identified.
- Sanitization: Absent.
Audit Metadata