The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify existing source code, creating a surface for indirect prompt injection if the processed code contains malicious instructions.
Ingestion points: The skill ingests project source code, including flow tracing and caller inspection as described in SKILL.md.
Boundary markers: The instructions contain explicit safeguards under the 'Complexity that must remain' section, directing the agent to preserve security, privacy, and authorization controls.
Capability inventory: The skill provides instructions to modify and delete code within the repository to reduce implementation complexity.
Sanitization: There are no technical sanitization or filtering mechanisms defined for the external content processed by the skill.
[NO_CODE]: The skill does not contain any executable scripts, binaries, or automated installation commands; it consists entirely of natural language instructions and a license file.

superior-lean