The Agent Skills Directory

[SAFE]: No malicious patterns, data exfiltration, or obfuscation were detected. The skill's functionality is strictly limited to local development and testing within the repository workspace.
[COMMAND_EXECUTION]: The skill utilizes shell commands to execute local Python modules, run pytest suites, and interact with Docker containers for database validation. These operations are restricted to the local environment and are standard for developer testing workflows.
[PROMPT_INJECTION]: The skill processes external URLs as input for its summarization tests, creating an indirect prompt injection surface. This is the intended purpose of the bot being tested.
Ingestion points: URLs provided to the CLI runner and content in simulated Telegram messages.
Boundary markers: The documentation mentions an internal pipeline (scraper chain to LLM) but does not define specific prompt delimiters.
Capability inventory: The toolchain performs network requests (scraper chain) and database writes (requests table).
Sanitization: The skill incorporates JSON contract validation and repair logic to handle potentially malformed or adversarial outputs from the LLM processing stage.

testing-workflows