The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses structured instructions for project estimation and contains no behavioral overrides. It processes project documents and user input for context. 1. Ingestion points: Reads documentation from .chalk/docs/ and user arguments. 2. Boundary markers: Absent. 3. Capability inventory: Uses the Write tool to create local markdown files. 4. Sanitization: No sanitization is performed on processed context. No malicious intent was observed.- [DATA_EXFILTRATION]: No network operations or sensitive file access patterns detected. File access is limited to local project documentation.- [REMOTE_CODE_EXECUTION]: No remote scripts, package installations, or external dependencies are involved.- [COMMAND_EXECUTION]: The skill utilizes restricted file-system tools (Read, Glob, Write, Grep) for documentation management within specified project paths.- [SAFE]: No security issues detected in the skill instructions or metadata.

create-estimation