The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The SKILL.md file contains a hardcoded database connection string example (mysql+pymysql://tools:tools-password@127.0.0.1:4306/tools_test) which includes an embedded username and password. While these are intended for a local development environment (localhost), hardcoding credentials in documentation is a poor security practice.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run tests, such as .venv/bin/python -m pytest. This capability allows the agent to execute code within the project's environment as part of its primary testing function.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from the project's source code to write and run tests, creating a potential surface for indirect prompt injection.
Ingestion points: Project source code files including controllers, models, workers, and routes (specified in SKILL.md).
Boundary markers: None are specified to differentiate between the skill's instructions and the content of the files being processed.
Capability inventory: The skill has the ability to execute shell commands via pytest and read/write files within the repository.
Sanitization: There is no evidence of sanitization or filtering applied to the source code content before it is processed by the agent.

flask-tests