Does this skill ask questions for every task?

No. It first attempts to resolve uncertainties by reading files and checking the environment. It only prompts the user for high-priority information that cannot be found automatically.

Can I exit the inquiry process early?

Yes, the protocol includes escape conditions and allow users to declare context as sufficient at any point during the inquiry phase.

What is the Aitesis protocol used in this skill?

Aitesis is an epistemic protocol that structures how AI identifies and resolves context insufficiency, ensuring it has 'informed execution' rather than acting on assumptions.

How does this skill reduce 'hallucinations' in AI coding?

By forcing the AI to scan for uncertainties and verify them against the codebase before execution, it prevents the AI from making incorrect guesses about your project structure.

Smart Context Inquiry

Name: Smart Context Inquiry
Author: jongwony

byjongwony

•

生産性とワークフロー

Ensures accurate AI execution by proactively identifying and resolving context gaps through structured inquiry before taking action.

Smart Context Inquiry implements the Aitesis protocol to sense context sufficiency before Claude executes a task. Instead of proceeding with assumptions, this skill scans your codebase and task requirements for uncertainties across factual, coherence, and relevance dimensions. It attempts to self-resolve gaps through environment observation and codebase exploration, surfacing only high-priority, information-rich questions to the user. This ensures that every action Claude takes is based on a complete and verified understanding of your project's state and intent.

主な機能

01Automated evidence collection via deep codebase and environment exploration

02Multi-dimensional uncertainty detection across factual, coherence, and relevance layers

03Convergence tracing to demonstrate exactly how context gaps were resolved

04Dynamic observation of environmental facts to reduce unnecessary human prompts

05Epistemic classification that prioritizes information-gain for user inquiries

0688 GitHub stars

ユースケース

01Resolving ambiguity in complex feature requests before generating code changes

02Identifying missing architectural context when working in large, unfamiliar codebases

03Validating environmental preconditions and dependencies before executing deployment scripts

主な機能

01Automated evidence collection via deep codebase and environment exploration

02Multi-dimensional uncertainty detection across factual, coherence, and relevance layers

03Convergence tracing to demonstrate exactly how context gaps were resolved

04Dynamic observation of environmental facts to reduce unnecessary human prompts

05Epistemic classification that prioritizes information-gain for user inquiries

0688 GitHub stars

ユースケース

01Resolving ambiguity in complex feature requests before generating code changes

02Identifying missing architectural context when working in large, unfamiliar codebases

03Validating environmental preconditions and dependencies before executing deployment scripts