The Agent Skills Directory

[SAFE]: The skill is purely instructional and descriptive. It does not contain any executable code, scripts, or calls to external APIs or tools. All operations occur within the text-based interaction with the agent.
[INDIRECT_PROMPT_INJECTION]: The skill explicitly asks for user-provided input in the form of contract clauses. While this creates an ingestion surface for potentially untrusted data, the skill possesses no capabilities (such as file writes or network operations) that could be abused via injection, making the risk negligible.
[NO_CODE]: The skill consists only of markdown instructions and metadata, with no associated scripts or dependencies to analyze.

clause-explainer