The Agent Skills Directory

[SAFE]: The skill facilitates standard software development practices like test generation and environment probing without suspicious or high-risk behavior.
[COMMAND_EXECUTION]: The instructions guide the agent to execute standard test runners (e.g., pytest, npm test) and a provided local Python script to identify the repository's technology stack.
[DATA_EXFILTRATION]: The skill reads project configuration files (e.g., package.json, pyproject.toml) to determine the testing environment. This analysis is conducted locally, and no network operations were detected that would exfiltrate this data.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external repository data, which is a potential surface for indirect instructions.
Ingestion points: The agent and the scripts/repo_test_probe.py script read project configuration and source files from the target repository.
Boundary markers: Absent; the skill relies on the agent's internal capability to differentiate between code and instructions.
Capability inventory: The agent has permissions to write files and execute local commands for the purpose of running tests.
Sanitization: No explicit sanitization or filtering of the ingested data is performed.

write-tests