mountain-nomad-bc/pacify-x/reasoning-utility-controller
Claude Code and agent skill by mountain-nomad-bc/pacify-x
- community
Checking that the SKILL.md file is still in the repository.
Decide whether another reasoning, retrieval, verification, or tool step has enough expected value. Use when the task requires this bounded operational control; keep it inactive otherwise.
Install
npx vskill@latest install mountain-nomad-bc/pacify-x/reasoning-utility-controllerSKILL.md
reasoning-utility-controller
Purpose
Decide whether another reasoning, retrieval, verification, or tool step has enough expected value.
Use when
- loop may continue
- budget/latency matters
- verification repeats
Do not use when
- mandatory safety check remains
Required inputs
- information gain
- unresolved requirements
- state change
- repetition
- cost
- risk
- budget
Outputs
- continue/stop/escalate
- utility
- reason
- next permitted action
Procedure
- measure information gain.
- measure hypothesis reduction.
- detect repeated patterns.
- estimate decision impact.
- compare value/cost/risk.
- stop/escalate when utility collapses.
Guardrails
- mandatory postconditions remain.
- token length alone is not decision.
- loop opens circuit.
Integration points
- budget controller
- tool-loop breaker
- planner
Failure modes to test
- bad utility calibration
- important cheap check skipped
Operational metrics
- cost per outcome
- unnecessary calls
- premature stops
Completion requirements
The skill is complete only when structured outputs are emitted, evidence references resolve, declared postconditions are checked, and the execution wrapper records the result. Any memory remains a candidate until framework promotion controls accept it.
Research basis
Runtime binding
- Family:
loop - Binding:
runtime.operational_controls.run_control - Activation: metadata is discoverable at startup; load this body only after selection.
- Effects: read-only control-plane analysis unless a separately admitted composed capability declares more.
Completion and evidence
Return a structured decision, reasons, outputs, and evidence references. Treat unresolved provenance, permissions, required inputs, or postconditions as a fail-closed result. Candidate memory, research, generated skills, and speculative work never become active without separate admission and evidence.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryChecking.px/skills/reasoning-utility-controller/SKILL.md
Popularity
- Repository stars
- 9
- 7-day trend score
- 17
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/mountain-nomad-bc/pacify-x/reasoning-utility-controller)