The Agent Skills Directory

[SAFE]: The skill functions as a structured reasoning assistant for developing test guides. It defines a persona ("The Oracle") and a clear step-by-step process for gathering requirements and documenting test criteria without performing dangerous actions.
[COMMAND_EXECUTION]: While the skill has access to the 'Bash' tool, it is primarily intended to support development workflows (e.g., verifying build commands). The instructions do not contain any patterns for executing arbitrary, obfuscated, or remote commands.
[DATA_EXFILTRATION]: No network activity or attempts to access sensitive system files (such as SSH keys or AWS credentials) were detected. The skill's file operations are limited to reading project specifications and writing documentation within the project's local directory.
[PROMPT_INJECTION]: The skill exhibits a standard surface for indirect prompt injection as it processes external files and user input to generate documentation.
Ingestion points: The skill reads the 'prompt.md' specification file and accepts interactive input via the 'AskUserQuestion' tool.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the ingested text are present.
Capability inventory: The toolset includes 'Bash', 'Write', and 'Read' permissions.
Sanitization: Content from the specification file is processed as descriptive text for documentation and is not specifically sanitized or escaped within the skill definition.

more-loop-oracle