The Agent Skills Directory

[DYNAMIC_EXECUTION]: The execution script scripts/run.py dynamically calculates the repository root by searching parent directories for a specific marker file (AGENTS.md) and appends this path to the Python module search path (sys.path). It then imports modules (tooling.common, tooling.quality_gate) from this computed path, which is categorized as dynamic loading from computed paths.
[INDIRECT_PROMPT_INJECTION]: The skill processes markdown files (S*.md) that are considered untrusted external data.
Ingestion points: The script reads content from files within the workspace/sections/ directory.
Boundary markers: No delimiters or safety instructions are used to wrap the processed text.
Capability inventory: The skill has the ability to read, write, and delete files within the provided workspace path.
Sanitization: There is no validation or sanitization of the input text before it is manipulated and written back to the file system.
[COMMAND_EXECUTION]: The SKILL.md file defines a command to execute a Python script using uv run. This allows the agent to run code that performs file system modifications within a specified workspace.

paragraph-curator