The Agent Skills Directory

[SAFE]: No malicious patterns, data exfiltration, or obfuscation techniques were identified in the skill instructions or the accompanying Python script.- [EXTERNAL_DOWNLOADS]: The skill utilizes the standard 'python-docx' library for Word file manipulation, which is a well-known and trusted package in the Python ecosystem.- [COMMAND_EXECUTION]: The skill executes a bundled Python script ('proofreader.py') to process documents. This execution is limited to local file operations (read/write) required for the skill's primary function and does not involve arbitrary shell execution.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from external docx files. Evidence: 1. Ingestion points: Paragraph text extracted via 'proofreader.py'. 2. Boundary markers: Absent. 3. Capability inventory: No network access or dynamic code execution capabilities are present to be exploited. 4. Sanitization: No sanitization of ingested document content is performed.

word-proofreader