The Agent Skills Directory

[SAFE]: No malicious behavior or patterns were detected. The skill's functionality is limited to formatting user-provided text into markdown templates and performing local filesystem directory listings to manage document numbering and file placement, which is consistent with its stated purpose as a documentation tool.
[INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection because it ingests untrusted user context and interpolates it into generated documents. However, this is the intended functionality, and the risk is minimal as the ingested data is not used in executable contexts or security-sensitive logic.
Ingestion points: The skill collects information through user prompts and an interactive workflow (Step 1) using the AskQuestion tool to define decision context, rationales, and consequences.
Boundary markers: The skill does not define specific delimiters or warnings to ignore instructions embedded within the user-provided context.
Capability inventory: The skill performs local directory scanning (e.g., docs/adr/, adr/) to determine file counts and writes markdown content to the local filesystem. It lacks network access or shell execution capabilities.
Sanitization: No explicit sanitization or filtering of user input is performed before interpolation into the ADR templates.

create-adr