The Agent Skills Directory

[SAFE]: The skill provides stylistic instructions for technical writing based on established frameworks (Di!taxis, Google Developer Style, and Simplified Technical English). The content is purely instructional and contains no malicious code, obfuscation, or persistence mechanisms.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted text (such as READMEs, PR descriptions, and RFCs), which technically creates a surface for indirect prompt injection. However, the risk is minimal as the skill performs no privileged actions.
Ingestion points: The skill is intended to review or write external content like docs, RFCs, and PR descriptions in SKILL.md.
Boundary markers: No specific delimiters or boundary markers are defined in the instructions.
Capability inventory: The skill contains no executable code, subprocess calls, network operations, or file-write capabilities. It is also restricted by the disable-model-invocation: true policy.
Sanitization: No sanitization of input data is performed, but the lack of capabilities limits the impact of potential injections.

technical-writing