aro-hcp-ev2-releases
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from HCP build logs via the
oncallCLI, which creates a theoretical surface for indirect prompt injection if build logs contain malicious instructions. - Ingestion points: The skill captures stdout from the
oncallCLI inscripts/get-ev2-releases.shandscripts/get-ev2-releases.ps1. - Boundary markers: Instructions in
SKILL.mddirect the agent to present the markdown table to the user "as-is" without additional delimiters. - Capability inventory: The skill uses the
shelltool to execute local scripts and the internaloncallCLI. - Sanitization: The scripts filter and format the CLI output into a markdown table but do not perform explicit sanitization of the content extracted from build logs.
Audit Metadata