The Agent Skills Directory

[SAFE]: No malicious patterns, prompt injections, or data exfiltration vectors were detected across the skill's instructions and reference materials.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided text for rewriting, which is an inherent attack surface for indirect prompt injection. The skill mitigates this risk through a rigorous evidentiary framework.
Ingestion points: Processes arbitrary text provided by users for drafting, rewriting, and adaptation as defined in the SKILL.md file.
Boundary markers: The references/fact-integrity.md file defines a Source Ledger to strictly classify and delimit supplied facts, author judgments, and unknown claims.
Capability inventory: The skill does not request or utilize any external tools or command execution capabilities in its frontmatter or logic.
Sanitization: Includes explicit instructions to preserve protected technical text (commands, paths, code) verbatim and to redact sensitive information such as secrets, tokens, or private keys if encountered in source text.
[NO_CODE]: The skill consists entirely of instructional markdown and YAML configuration files; no executable scripts or binaries are distributed with the skill.

human-writing