The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Bash commands for routine development tasks such as file discovery (ls, grep), path validation (test), and version control operations (git commit). These operations are consistent with the skill's stated purpose of managing a test-addition workflow.
[INDIRECT_PROMPT_INJECTION]: The skill establishes a workflow that ingests content from design documents and UI specifications to serve as context for subagents. While this is a standard operational surface for an orchestrator, it is documented here as an inherent architectural property.
Ingestion points: Document paths provided via $ARGUMENTS and existing markdown files in docs/design/ and docs/ui-spec/.
Boundary markers: Absent; the skill interpolates discovered paths directly into prompts for subagents like acceptance-test-generator without explicit delimiters.
Capability inventory: Includes file system access via Bash, file creation (task files), and source code modification through Git commits.
Sanitization: The skill does not perform validation or filtering on the content of the documents before they are processed by subagents.

recipe-add-integration-tests