The Agent Skills Directory

[SAFE]: The skill is designed to manage project documentation, specifically CONTEXT.md and ADR (Architecture Decision Record) files. Its operations are restricted to reading existing documentation and writing updates to clarify domain language.
[SAFE]: No evidence of network activity, data exfiltration, or hardcoded credentials was found. The skill does not interact with external APIs or non-local domains.
[SAFE]: The skill does not execute shell commands, install packages, or download remote scripts. It uses simple file-based logic for documentation formatting.
[SAFE]: There are no signs of obfuscation (such as Base64 or hidden characters) or prompt injection attempts designed to bypass agent safety guidelines.
[SAFE]: While the skill ingests untrusted project data (Indirect Prompt Injection surface), the risk is negligible as the skill is only authorized to update text-based documentation.
Ingestion points: Reads project files CONTEXT.md, CONTEXT-MAP.md, and ADRs.
Boundary markers: Absent.
Capability inventory: Limited to reading and writing markdown documentation files.
Sanitization: Absent.

clarify-domain-docs