The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes data from external primary sources such as official documentation, source code, and RFCs. While this constitutes an ingestion surface for potential indirect prompt injection, the behavior is core to the skill's purpose, and instructions focus on citing and reporting rather than executing external content. 1. Ingestion points: Technical documentation, source code, and RFCs accessed during the research process. 2. Boundary markers: The instructions do not define specific delimiters to isolate external content from the agent's instructions. 3. Capability inventory: The skill identifies research questions and writes structured findings to markdown files within the .lcs/work-items/ directory. 4. Sanitization: No explicit sanitization or filtering of research content is specified in the behavior checklist.

lcs-research