unblock
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): No malicio us pattern s detecte d. The files consist of markdow n instructio ns for guiding an AI agent throug h standar d softwar e developmen t workflow s.
- [Indirec t Promp t Injectio n] (LOW): The skill describe s proces s ing externa l data such as PR histor y, erro r message s, and ticke t URLs. Evidence chain: 1. Ingestio n point s: PR s and ticke t s via
data_retrieva landlink_resolve r. 2. Boundar y marker s: No ne specifie d in the referenc e docs. 3. Capabilit y inventor y: Contex t querie s viaunblocke d_contex t_engine. 4. Sanitizatio n: No ne specifie d. The risk is considere d LOW as the se are procedura l temlate s rathe r than executable code.
Audit Metadata