react-state
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The
SKILL.mdfile contains an 'Agent Workflow (MANDATORY)' section that uses imperative language to override the agent's autonomous behavior. It dictates specific sub-agent spawning ('spawn 3 agents in parallel') and tool usage ('fuse-ai-pilot:sniper') as a mandatory sequence of operations. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves analyzing and modifying a project's codebase, which creates a surface for instructions embedded in external files to influence the agent's actions.
- Ingestion points: The local project codebase, which is analyzed via the
fuse-ai-pilot:explore-codebasetool as specified inSKILL.md. - Boundary markers: None identified; the instructions do not specify delimiters or warnings to ignore instructions found within the analyzed codebase.
- Capability inventory: The agent is authorized to implement global state management (writing/modifying files) and execute validation tools like
fuse-ai-pilot:sniperas described inSKILL.md. - Sanitization: No evidence of sanitization, filtering, or validation of the ingested codebase content is present in the skill instructions or references.
Audit Metadata