setup-and
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs repository organization tasks such as moving files using
git mvand potentially executing backend-specific operations for managing GitHub labels and issues. - [EXTERNAL_DOWNLOADS]: The skill includes a command to install missing dependencies:
npx --yes skills add mattpocock/skills. This involves downloading and executing code from the NPM registry and a public GitHub repository. This risk is mitigated by the instruction that installation should only occur upon explicit user request and approval within a write envelope. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes instructions from files within the repository environment.
- Ingestion points: The skill reads instructions from
.and/config.yml,AGENTS.md, andCLAUDE.mdin the target repository to determine setup actions. - Boundary markers: No explicit delimiters are used to isolate content read from files from the agent's core instructions.
- Capability inventory: The skill possesses the capability to modify the file system, execute git commands, and run npx-based installation scripts.
- Sanitization: While the skill validates the schema of the YAML configuration, there is no validation or sanitization for natural language instructions parsed from the markdown files.
Audit Metadata