Route agent actions through a predefined skill routing table before execution.
npx skills add https://github.com/Rune-kit/rune --skill skill-router-rune-kit
Please help me install this Agent Skill. Skill: skill-router Source: https://github.com/Rune-kit/rune/tree/main/skills/skill-router Command: npx skills add https://github.com/Rune-kit/rune --skill skill-router-rune-kit
This Skill acts as a mandatory enforcement layer, ensuring that every agent action involving code, files, or technical decisions is routed through the correct, pre-defined skill, preventing unauthorized actions and enforcing a disciplined check-before-act protocol.
CODE_CHANGE, QUESTION, DEBUG_REQUEST, etc., to determine enforcement levels.rune:debug, then to rune:fix, and finally to rune:test, preventing the agent from attempting to fix the bug without proper debugging or testing procedures.Use the skill-router to ensure all code-related requests are routed to the appropriate skill before execution.
High-intent search queries and answers about installing and using this skill.
To enforce skill routing for AI actions, you need a meta-enforcement layer that intercepts requests and routes them through a defined skill routing table before any code or file decisions are made. This prevents unauthorized actions.
Intent classification in agent workflow control categorizes user requests into types like CODE_CHANGE, QUESTION, or DEBUG_REQUEST to determine the appropriate enforcement level and route the action to the correct skill.
To prevent AI agents from rationalizing skipping workflows, implement a mandatory check-before-act protocol that enforces routing discipline and prevents the agent from bypassing pre-defined technical decision procedures.
Yes, skill routing supports adaptive routing overrides and file ownership constraints, allowing you to maintain enhanced workflow integrity while adjusting enforcement rules for specific agent actions.
You need a mandatory routing layer for AI agent requests when processing complex multi-step prompts, such as asking an agent to fix a bug and add tests, ensuring each phase routes sequentially to the correct debug, fix, and test skills.
Yes, you can route a single request through multiple skills sequentially by classifying the intent and enforcing a routing table, ensuring a debug request routes to the debug skill before proceeding to the fix and test skills.
William189189Summarize the COC five-layer architecture and anti-amnesia framework for AI-assisted development.
SimHackerSystematically debug issues by hypothesizing, testing, and tracking evidence.
oozoofrogRoute iOS/Swift tasks to appropriate sub-skills using a routing guide.
phuonghxRoute incoming requests to specialist agents based on content analysis.