The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Node.js scripts to manage the ADR lifecycle. It utilizes scripts/resolve-feature.js to determine the target directory and skills/adr/scripts/next-adr-number.js to compute the next available sequence number. These operations are performed using the restricted Bash(node:*) tool.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a documentation generator that ingests user-provided text to populate architectural records.
Ingestion points: ADR metadata (Title) and descriptive sections (Context, Decision, Consequences, Alternatives) gathered during Phase 3.
Boundary markers: The skill does not employ specific delimiters or "ignore instructions" warnings for the interpolated markdown content.
Capability inventory: The skill uses Write and Edit tools to store the records in the docs/features/ directory.
Sanitization: The skill includes a resolution gate that validates the feature slug using a case-insensitive regex pattern (/^[a-z0-9][a-z0-9._-]*$/i) to prevent path traversal attempts such as ../evil.

adr