antigravity-skill-orchestrator
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a master catalog from the author's GitHub repository at
https://raw.githubusercontent.com/sickn33/agentic-awesome-skills/main/CATALOG.md. This is identified as a vendor-owned resource used for discovery of curated skills. - [PROMPT_INJECTION]: The skill includes instructions to evaluate task complexity and prevent the use of specialized skills for simple tasks. These are functional guardrails designed to improve efficiency rather than bypass safety filters.
- [COMMAND_EXECUTION]: The instructions explicitly forbid the agent from creating or generating new skills at runtime, restricting behavior to the combination of existing, verified skills.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from a skill catalog to identify appropriate tools.
- Ingestion points: SKILL.md Step 3.3 (Fetching external
CATALOG.mdvia web/CLI tools). - Boundary markers: Absent.
- Capability inventory: Memory tool access (
memory_search,memory_write) and environment-specific terminal/file capabilities. - Sanitization: Absent. The surface is used for metadata discovery from a vendor-controlled source, posing no immediate threat.
Audit Metadata