The Agent Skills Directory

[SAFE]: The skill implements a set of context engineering tools and templates without any detected malicious patterns or security risks.
[COMMAND_EXECUTION]: The skill utilizes local shell scripts (detect_state.sh, analyze_structure.sh, and estimate_tokens.sh) to perform codebase analysis. These scripts use standard Unix utilities such as find, grep, and wc to inspect directory structures and file sizes. The scripts follow safe practices, including the use of set -e and proper argument handling.
[DATA_EXFILTRATION]: No network-enabled commands or exfiltration patterns were identified. All operations, including repository scanning and token estimation, are confined to the local project environment.
[PROMPT_INJECTION]: The instructions in SKILL.md are focused on a structured documentation workflow (Detect, Route, Measure, Decide, Execute, Maintain) and do not contain attempts to override system prompts or bypass safety guidelines.

intent-layer