The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external templates (templates/sales.json) and an external agent registry to construct prompts, creating a surface for potential injection.
Ingestion points: templates/sales.json, ruflo's agent registry (referenced in SKILL.md).
Boundary markers: Absent; no delimiters or "ignore instructions" are specified for the constructed prompts.
Capability inventory: File system access via Bash for writing ledgers and BBS logs, and command execution for running simulation scripts.
Sanitization: Partial; while schema validation is mentioned for the pod template, no sanitization or escaping is described for the prompt construction process.
[EXTERNAL_DOWNLOADS]: The skill resolves agent configurations against an external entity called "ruflo's agent registry."
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute the pod-tick.mjs script and manage file-based budget and status stores.

pod-sales