The Agent Skills Directory

[SAFE]: The skill's primary function is prose editing and stylistic refinement. It provides detailed registers and checklists for Technical, Editorial, and Correspondence writing, ensuring high-quality output without bypassing safety filters or security protocols.\n- [COMMAND_EXECUTION]: The skill includes a local utility script, scripts/prose_audit.py, intended for mechanical prose auditing. The script uses standard Python libraries and regular expressions to analyze text files provided by the user or agent. It does not perform network operations, access sensitive system files, or execute arbitrary commands.\n- [PROMPT_INJECTION]: As a writing and editing tool, the skill processes user-provided drafts. The instructions provide specific workflows for inventorying facts and transforming structure, such as the 'fact inventory' step, which effectively mitigates the risk of the agent being misled by embedded instructions within user-provided text drafts.\n- [DATA_EXFILTRATION]: No hardcoded credentials, sensitive file paths, or network exfiltration patterns were found. Network operations are absent from the included script and instruction files.\n- [SAFE]: No obfuscated code or hidden instructions (Base64, Unicode homoglyphs, or zero-width characters) were detected. All content is transparently documented and readable.

antislop-writing