The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute specific Python scripts (skill_validator.py, script_tester.py, and quality_scorer.py) via shell commands. These scripts are referenced as local dependencies but are not included in the provided skill files, making their implementation unverifiable.
[REMOTE_CODE_EXECUTION]: The tool script_tester.py is described as a utility that executes other Python scripts found within a user-specified directory to validate runtime behavior. This constitutes execution of code from computed, potentially untrusted paths.
[PROMPT_INJECTION]: The skill's primary function is to ingest and analyze external SKILL.md files and associated scripts, including those designed as adversarial test cases. This creates a surface for indirect prompt injection where malicious instructions embedded in the analyzed data could influence the agent's actions.
[PROMPT_INJECTION]: Mandatory Evidence Chain for Indirect Prompt Injection:
Ingestion points: Target SKILL.md and Python script files located at the path provided during execution.
Boundary markers: Absent; the instructions do not provide any delimiters or negative constraints to prevent the agent from following instructions found within the files it is testing.
Capability inventory: Execution of shell commands and reading of arbitrary local files.
Sanitization: None; the agent is not instructed to sanitize or escape the content of the files before processing or simulating their behavior.

skill-tester