Route BAZINGA workflow transitions to the next agent via database state-machine rules.
npx skills add https://github.com/mehdic/CDC --skill workflow-router
Please help me install this Agent Skill. Skill: workflow-router Source: https://github.com/mehdic/CDC/tree/main/.claude/skills/workflow-router Command: npx skills add https://github.com/mehdic/CDC --skill workflow-router
It decides the next step in the BAZINGA workflow by consulting the seeded transitions and applying testing, escalation, and security rules to determine the next agent and action.
After an agent responds, invoke this skill to determine the next actor (e.g., developer -> qa_expert) and whether to spawn or respawn.
High-intent search queries and answers about installing and using this skill.
Workflow routing determines the next agent in a sequence by consulting state-machine transitions stored in a database. After any agent responds, the router applies the seeded transitions and rules to output the next target agent, action type, and contextual flags for spawning or respawning.
You need an initialized bazinga/bazinga.db with pre-seeded state-machine transitions and configuration rules. The router reads these transitions to make routing decisions, so the database must be populated with valid transition paths and any testing, escalation, or security override rules before invoking the skill.
Yes. Workflow routing applies testing_mode, escalation, and security overrides when determining the next action. These special rules modify routing behavior based on context, allowing you to bypass normal transitions during testing or escalate to different agents when security conditions are met.
The router outputs a structured JSON object describing the routing decision, including the next_agent to spawn, the action type to perform, the model to use, and contextual flags. This output directly feeds the next step in your BAZINGA workflow.
Yes. Workflow routing operates after an agent has responded. It takes the latest agent response as input, consults the state-machine transitions, and determines which agent should act next based on the current workflow state and configured rules.
Snowflake-LabsA concise summary of your roleplay character: name,</name> personality, and/or behavior, formatted as a single string sentence. Typically不超过50字，此处省略
spiral-app-labsDetect broken Codex ACP subagents and trigger a safe respawn workflow.
jpoutrinAutomate QA test file naming, directory structure, and status transitions.
pazdedavDefine DAG workflow graphs with JSON schema for agent orchestration.
