tailwindcss-accessibility
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses authoritative directives to override the agent's default planning and orchestration. The instructions mandate a specific parallel sub-agent invocation workflow ("MANDATORY", "Before ANY implementation, spawn 3 agents in parallel") using specific named sub-agents.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a multi-stage execution pipeline involving research and validation agents, creating a vulnerability surface for cross-agent instruction leakage.
- Ingestion points: Existing codebase patterns and research results (SKILL.md).
- Boundary markers: Absent; no delimiters are used to separate skill instructions from the context generated by sub-agents.
- Capability inventory: The workflow includes spawning sub-agents (
fuse-ai-pilot) and querying context servers (mcp__context7__query-docs). - Sanitization: No explicit data sanitization or instruction filtering is defined for the information flow between the exploration, research, and validation ("Sniper") stages.
Audit Metadata