legal-text-format
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes legal texts provided by users or external tools like wechat-article-fetch to perform structure identification and formatting. 1. Ingestion points: Input texts are received via user paste, local files, or chain-called skills as described in SKILL.md. 2. Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands within the processed legal content. 3. Capability inventory: The skill performs file system write operations to an archive directory using scripts/format_legal_cases.py. 4. Sanitization: The formatting script employs extensive regex-based cleaning and punctuation normalization to enforce structure.
- [COMMAND_EXECUTION]: The test file scripts/tests/test_format_legal_cases.py contains calls to subprocess.run(). These calls are used to verify the CLI functionality of the formatting script by executing it with controlled local inputs and outputs during the testing phase, and they do not form part of the primary runtime logic for text formatting.
Audit Metadata