The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (e.g., dae_resolve.py, dae_progress.py, dae_branch.py) from the ${PLUGIN_ROOT}/scripts/ directory to manage feature initialization.
[COMMAND_EXECUTION]: Shell commands are used for Git operations (git checkout -b <slug>). The skill validates the <slug> input (kebab-case, ASCII, max length) to prevent command injection vulnerabilities.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes data from user interviews, context payloads, and the existing codebase to generate feature documentation.
Ingestion points: feature_intake payload, user interview fields, and existing project metadata (SKILL.md).
Boundary markers: None specified for the output feature.md file.
Capability inventory: Subprocess execution for local Python scripts and Git commands (SKILL.md).
Sanitization: While the slug is strictly validated, other free-text fields like title and scope are written to files without explicit sanitization.
[SAFE]: No obfuscation, hardcoded credentials, or unauthorized network exfiltration was detected. All external references are to well-known documentation services (Notion).

feature-init