v4-hook-generator
Warn
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill's YAML frontmatter lists 'uniswap' as the author, which conflicts with the actual provider information ('fuzzystodd'). This misattribution is deceptive and may cause users to trust the skill as an official tool when it is not verified as such.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for gathering user requirements to generate financial contract code, creating a surface for indirect prompt injection attacks.
- Ingestion points: User-provided specifications for hook logic, permissions, and configuration options gathered during the 'Gather Requirements' step.
- Boundary markers: The instructions lack delimiters or specific instructions to the agent to disregard instructions potentially embedded within the user's requirements.
- Capability inventory: The skill is permitted to use Bash and WebFetch, and it specifically facilitates calling the 'generate_hook' tool to produce Solidity source code.
- Sanitization: There are no instructions or mechanisms to validate, filter, or sanitize the user-supplied data before it is interpolated into the MCP tool call or the generated output.
Audit Metadata