The Agent Skills Directory

[SAFE]: The skill defines a legitimate procedural workflow for software development. It instructs the agent to read planning artifacts and implement code changes based on them.
[DATA_EXFILTRATION]: No sensitive data access or network exfiltration patterns were detected. The skill only interacts with project-specific documentation and source code.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from project files to influence agent actions. However, this risk is inherent to the primary function of a coding assistant.
Ingestion points: tasks.md, design.md, and specs/ (identified in SKILL.md).
Boundary markers: Not specified in the instructions.
Capability inventory: Reading and writing files, modifying code, and executing tests (identified in SKILL.md).
Sanitization: Not specified in the instructions.

openspec-apply