بنقرة واحدة
confidence-calibrator
Heuristics to determine when to act automatically vs when to ask for permission.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Heuristics to determine when to act automatically vs when to ask for permission.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | confidence-calibrator |
| description | Heuristics to determine when to act automatically vs when to ask for permission. |
"Know what you don't know."
Condition: "I have seen this pattern 100 times, I have verified the file exists, and I understand the inputs/outputs perfectly." Action: Proceed. (Execute, Edit, Run).
Condition: "I know the general approach, but there are two valid ways (e.g., library A vs B), or the code is legacy/brittle." Action: Propose Options.
Condition: "User asked for 'optimization' but didn't define metrics," or "I'm guessing the function name." Action: STOP & ASK.
Enforces modularity, reusability (DRY), and high-level architectural patterns. Focuses on SOLID principles, Design Patterns, and Scalability.
Universal architectural auditor for any tech stack. Enforces Separation of Concerns, Clean Architecture, and best practices by identifying the project structure and detecting layer violations.
Defines rules for skill activation, including input requirements, output contracts, conflict lists, and performance costs. Enforces Lean Execution mode by default.
Implementation of Enterprise patterns: CQRS, Event Sourcing, Domain-Driven Design (DDD).
Generates robust, production-grade 3rd party API integrations with Retry, Rate Limiting, and DTOs.
Workflow for deeply researching and architecting a new module/feature. STRICTLY NO CODE EXAMPLES. Use this for high-level technical design and research before implementation.