The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, such as tool descriptions, error messages, and inter-agent instructions, for the purpose of simplification and rewriting. This creates a surface where malicious instructions embedded in the input text could potentially influence the agent's behavior.
Ingestion points: The skill reads input English text for meaning as part of the simplification process (SKILL.md, README.md).
Boundary markers: No specific delimiters or boundary markers (such as XML tags or explicit 'ignore instructions' blocks) are mandated in the instructions for isolating the untrusted input text.
Capability inventory: The agent is authorized to execute a local Python script (scripts/ste-lint.py) and perform file-write operations to deliver simplified documentation.
Sanitization: There are no instructions for sanitizing or escaping the input content before the agent processes it.
[COMMAND_EXECUTION]: The skill's workflow involves the execution of a local Python utility (scripts/ste-lint.py) to perform deterministic linting of structural STE rules. While the script uses only standard libraries and is provided as part of the skill package, it constitutes a deliberate shell execution step.
[EXTERNAL_DOWNLOADS]: The documentation (README.md) provides installation instructions that involve downloading the skill package from an external GitHub repository (danyuchn/asd-ste100-skill) using the npx skills CLI tool.

clear-writing