kenoodl-synthesis
kenoodl-synthesis 제작
Web & Frontend Development요약
kenoodl is an orthogonal injection engine — it synthesizes patterns from domains outside your agent's training data.
설명
이 스킬이 하는 일
kenoodl maps an agent's working context against pattern libraries from outside its training data and returns novel structure. Use it when agent outputs start repeating or converging on the same solutions. Context is processed and discarded — stateless by design, nothing stored or used for training.
Additional prompting or compute can't escape the model's own embedding space, but a single kenoodl call injects structure from unrelated domains that the model couldn't generate on its own.
사용 시점
- Breaking an agent loop stuck generating variations of the same output
- Injecting cross-domain patterns into a stalled reasoning or brainstorming session
- Reducing token burn from circular generation in long autonomous runs
- Getting a new structural framing when a problem-solving approach has plateaued
- Adding external signal to autonomous agent loops without requiring human input
예시 워크플로
User asks: My agent keeps producing slight variations of the same solution. Break the loop.
- 1
Detect that output entropy has dropped across recent iterations
- 2
Collect the current working context — problem statement, prior outputs, and constraints
- 3
POST context to kenoodl.com/api/cli/drop with Bearer KENOODL_TOKEN authorization
- 4
Poll GET /api/cli/status/<jobId> every 3 seconds until synthesis completes
- 5
Inject the returned cross-domain structure back into the agent's working context
Novel structured signal from an unrelated domain, giving the agent a new starting point
요구 사항
- KENOODL_TOKEN environment variable
- kenoodl.com account to generate a CLI token
- USDC on Base blockchain for x402 autonomous access path ($0.49 per call, 20/day per wallet)
설치
clawhub install kenoodl-synthesis/kenoodl-synthesisCLI 명령을 실행하거나, 프롬프트를 Claude Code, Cursor, Codex 등 사용하는 코딩 에이전트에 붙여넣으세요.