The Agent Skills Directory

[SAFE]: The skill implements logic for text parsing and risk assessment using standard Python libraries (pandas, re, datetime). No suspicious network operations or hardcoded credentials were found.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted contract text, creating a surface for indirect prompt injection. However, the risk is minimized as the data is processed via fixed regular expressions rather than being directly re-interpolated into executable instructions or sensitive sinks.
Ingestion points: Contract text is ingested via the analyze_text method in SKILL.md and read from local files in the provided examples.
Boundary markers: No explicit LLM boundary markers are used in the Python script, but the analysis logic is strictly confined to regex-based extraction.
Capability inventory: The skill possesses filesystem write capabilities via the export_analysis method in SKILL.md, which uses pandas to generate Excel workbooks.
Sanitization: The _analyze_clause method truncates the processed text to 500 characters, providing a basic limit on the volume of untrusted content stored in analysis results.

contract-clause-analyzer