skills/aeonfun/aeon/install-skill/Gen Agent Trust Hub

install-skill

Warn

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: MEDIUMPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The skill is designed to automatically merge code changes into the main branch using gh pr merge --auto. This allows the agent to modify the authoritative repository state without direct human intervention.
  • [COMMAND_EXECUTION]: The skill executes several powerful CLI tools, including git, gh (GitHub CLI), and a custom installation script bin/install-skill-pack. These tools are used to manage branches, push code, and interact with repository infrastructure.
  • [EXTERNAL_DOWNLOADS]: The skill downloads skill packs from external GitHub repositories (via codeload.github.com). While GitHub is a well-known service, fetching and integrating code from arbitrary repositories introduces a significant supply chain risk surface.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion point for external data by fetching skill packs. The instructions include a mandatory evidence chain: 1) Ingestion point is the bin/install-skill-pack call fetching the external repo. 2) Boundary markers include the landing of skills in a disabled state and the explicit instruction to never follow instructions found inside the fetched pack's files. 3) Capability inventory includes file system writes, branch management, and PR merging. 4) Sanitization is provided by a scripts/skill-scan.sh scanner.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 9, 2026, 03:35 PM