The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill integrates with external Python scripts located in the vendor's directory (${CLAUDE_PLUGIN_ROOT}/vendor/c-level-advisor/) to perform contract risk scanning and term sheet analysis. These scripts are executed via the command line with file paths provided as arguments.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and process untrusted external data (contracts and term sheets).
Ingestion points: Processes external files provided via path/to/contract.txt and path/to/term_sheet.json.
Boundary markers: No explicit boundary markers or 'ignore embedded instructions' warnings are provided in the command-line examples.
Capability inventory: The skill uses shell-based Python execution to process input files.
Sanitization: No evidence of sanitization or validation of the input file content is present in the skill instructions.

gc-review