The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests untrusted data from multiple documentation sources to guide its code generation and command execution.
Ingestion points: Reads program-design-{id}.md, task-breakdown-{id}.md, and unit-testing-{id}.md as specified in SKILL.md.
Boundary markers: None identified in the prompt templates to distinguish between instructions and data.
Capability inventory: The skill has broad capabilities including Bash(*), Write, and Read (defined in SKILL.md).
Sanitization: No sanitization or validation of the ingested documentation content is performed before use.
[COMMAND_EXECUTION]: The skill provides templates for executing arbitrary shell commands via Bash(*) to run tests and linters.
Evidence: templates/development-checklist.md contains command references for ./gradlew test, mvn test, pytest, and npm test.

code-development