agent-eval
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references a third-party repository (
github.com/joaquinhuigomez/agent-eval) for installation and source review. - [COMMAND_EXECUTION]: The tool's primary purpose is to execute arbitrary bash commands (e.g.,
pytest,npm run build,grep) defined in YAML task files to evaluate coding agents. - [INDIRECT_PROMPT_INJECTION]: The skill processes external YAML task definitions that combine instructions for agents with executable evaluation logic.
- Ingestion points: YAML task files located in the
tasks/directory. - Boundary markers: None identified; the instructions do not provide delimiters to separate untrusted task content from the agent's core instructions.
- Capability inventory: Uses
Bash,Write, andEdittools to perform file modifications and run evaluation commands. - Sanitization: None identified; there is no evidence of filtering or validation for the commands or prompts defined in the YAML configurations.
Audit Metadata