The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to collect and summarize results from subagents, creating a surface where untrusted data from background tasks is ingested into the main conversation context.
Ingestion points: Results from subagent tasks are retrieved for inspection and summary (SKILL.md, Step 6).
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the ingested content.
Capability inventory: The agent utilizes subagent orchestration tools to launch and manage tasks.
Sanitization: No explicit output sanitization is mandated, although naming rules for task references provide a basic level of input validation.
[EXTERNAL_DOWNLOADS]: The documentation references installation commands using the vendor's infrastructure at skills.flc.io and the author's GitHub repository.
[NO_CODE]: The skill consists entirely of Markdown and YAML configuration files. It does not include Python scripts, Node.js code, or other executable binaries.

async