The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill explicitly instructs the agent to follow [LOOP_SUGGESTION] and [CRON_SUGGESTION] hints provided in MCP tool responses. This creates a vulnerability where data returned from external tools (which might be influenced by external content) could direct the agent's actions.
Ingestion points: Data is ingested via the mcp__plugin_ruflo-core_ruflo__hooks_worker-dispatch and worker-status tools as referenced in SKILL.md.
Boundary markers: The instructions do not define boundary markers or delimiters to isolate these hints from other tool output or to verify their integrity.
Capability inventory: The skill possesses the capability to execute shell commands via Bash, dispatch workers via MCP, and persist sessions using ScheduleWakeup (per the frontmatter in SKILL.md).
Sanitization: There is no evidence of sanitization, filtering, or validation for the hints retrieved from the tool outputs.

loop-worker