gauntlet-loop-brain
Evidence-gated Obsidian brain for bounded multi-agent improvement loops.
파일 탐색기
- marketplace.json
- plugin.json
- bug_report.yml
- config.yml
- ci.yml
- CODEOWNERS
- dependabot.yml
- PULL_REQUEST_TEMPLATE.md
- external-evidence-over-self-description-curator.md
- gauntlet-loop-secretary.md
- honest-stopping-and-acceptance-curator.md
- one-initiating-prompt-many-governed-iterations-curator.md
- parallel-only-when-independent-curator.md
- loop-protocol.svg
- brain.css
- app.json
- appearance.json
- community-plugins.json
- core-plugins.json
- graph.json
- .gitkeep
- .manifest.json
- brain-relationship-map.svg
- decision.md
- source-note.md
- standard-note.md
- _index.md
- _index.md
- main.canvas
- Onboarding Canvas.canvas
- _index.md
- Best Practices Kernel.md
- Blind Comparison Bias.md
- Concrete Reference.md
- Coupling-Aware Fan-Out.md
- Evaluator Optimizer.md
- Feedback Memory.md
- Fresh Context Critic.md
- Iterative Refinement.md
- Judge Calibration.md
- One Prompt Is Not One Completion.md
- Protected Graders.md
- Self-Evaluation Limits.md
- Stop and Escalation Policy.md
- _index.md
- Approval Queue.md
- _index.md
- Action Roadmap.md
- Health Scorecard.md
- _index.md
- _index.md
- AI Marketing Hub pilot one opt-in skill with deterministic gates and no external.md
- Blind-order probe run each pairwise judgment in both A-B and B-A order.md
- Coupling probe parallel ownership versus sequential single-owner work on the sam.md
- Critic independence probe same-context, fresh-context, cross-model, and calibrat.md
- Equal-budget comparison one-pass agent, iterative single agent, and governed Gau.md
- Plateau probe quality gain and regression rate across increasing iteration budge.md
- Source To Claim Spot Check Probe.md
- _index.md
- AI Marketing Hub Adapter.md
- Blind Comparison and Judge Calibration.md
- Budget Plateau and Human Escalation.md
- Builder Critic Evidence Loop.md
- Claim Verification Flow.md
- Context Compaction Routine.md
- Coupling-Aware Decomposition.md
- Explore Plan Code Commit.md
- Gauntlet Fit and Reference Gate.md
- Integration Regression and Smoothing.md
- Multi-Agent Fan-Out Research Flow.md
- One-Prompt Job Contract.md
- Reporting Workflow.md
- Research Refresh Workflow.md
- Source Intake Workflow.md
- Synthesis Workflow.md
- _index.md
- AI Marketing Hub domain adapters and acceptance metrics have not yet been implem.md
- Evidence Coverage Not Yet Verified.md
- No controlled equal-budget benchmark yet isolates the Gauntlet prompt from model.md
- No general method guarantees a model critic can observe dynamic feel, long video.md
- Same-model fresh-context criticism reduces shared history but does not guarantee.md
- The best production plateau rule across domains remains unverified.md
- Answer Engine Citability Policy.md
- CONVENTIONS.md
- Corpus Scope Policy.md
- dashboard.md
- Memory Governance Policy.md
- Provenance Trace Policy.md
- Start Here.md
- Tag Taxonomy.md
- Uncertainty Eval Policy.md
- _index.md
- _index.md
- How should coupling be measured before fan-out.md
- What Current Official Source Resolves The Highest Risk Claim.md
- What equal-budget baseline would show that the loop adds value beyond more infer.md
- What evidence is sufficient for dynamic artifacts such as games, video, and inte.md
- When should a retry use the same critic, a fresh critic, multiple critics, or a.md
- Which dimensions require deterministic tests, model judges, domain experts, or r.md
- Which reference set best represents the intended audience and task rather than m.md
- _index.md
- Weekly Report.md
- _index.md
- Gauntlet Loop Evidence Baseline.md
- Source Manifest Guide.md
- hot.md
- hot.template.md
- index.md
- log.md
- log.template.md
- overview.md
- CODEX.md
- README.md
- shipping-rules.md
- gauntlet-loop-hero.png
- README.md
- AI_MARKETING_HUB_INTEGRATION.md
- GAUNTLET_JOB_CONTRACT.md
- MARKET_VALIDATION_PROTOCOL.md
- OPERATOR_KIT.md
- PRODUCT_BOUNDARIES.md
- PUBLIC_READINESS_AUDIT.md
- repository-metadata.md
- equal-budget-scorecard.md
- interview-guide.md
- retention-ledger.template.csv
- brain.css
- app.json
- appearance.json
- community-plugins.json
- core-plugins.json
- graph.json
- .gitkeep
- sample-source.md
- .manifest.json
- brain-relationship-map.svg
- decision.md
- source-note.md
- standard-note.md
- _index.md
- _index.md
- main.canvas
- Onboarding Canvas.canvas
- _index.md
- Best Practices Kernel.md
- Blind Comparison Bias.md
- Building effective agents.md
- Claude of Duty original prompt.md
- Claude of Duty README and honest assessment.md
- Concrete Reference.md
- Coupling-Aware Fan-Out.md
- Demystifying evals for AI agents.md
- Effective harnesses for long-running agents.md
- Evaluator Optimizer.md
- Feedback Memory.md
- Fresh Context Critic.md
- How to Run a Gauntlet Loop.md
- How we built our multi-agent research system.md
- Iterative Refinement.md
- Judge Calibration.md
- Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena.md
- One Prompt Is Not One Completion.md
- Our framework for developing safe and trustworthy agents.md
- Protected Graders.md
- Reflexion Language Agents with Verbal Reinforcement Learning.md
- SELF-[IN]CORRECT LLMs Struggle with Discriminating Self-Generated Responses.md
- Self-Evaluation Limits.md
- Self-Preference Bias in LLM-as-a-Judge.md
- Self-Refine Iterative Refinement with Self-Feedback.md
- Specification gaming the flip side of AI ingenuity.md
- Stop and Escalation Policy.md
- Towards a science of scaling agent systems When and why agent systems work.md
- _index.md
- Approval Queue.md
- _index.md
- Action Roadmap.md
- Health Scorecard.md
- _index.md
- Fixture gauntlet job.md
- Report Renderer render brain report.md
- Synthesis Module contract.md
- Test test gauntlet contract.md
- _index.md
- AI Marketing Hub pilot one opt-in skill with deterministic gates and no external.md
- Blind-order probe run each pairwise judgment in both A-B and B-A order.md
- Coupling probe parallel ownership versus sequential single-owner work on the sam.md
- Critic independence probe same-context, fresh-context, cross-model, and calibrat.md
- Equal-budget comparison one-pass agent, iterative single agent, and governed Gau.md
- Plateau probe quality gain and regression rate across increasing iteration budge.md
- Source To Claim Spot Check Probe.md
- _index.md
- AI Marketing Hub Adapter.md
- Blind Comparison and Judge Calibration.md
- Budget Plateau and Human Escalation.md
- Builder Critic Evidence Loop.md
- Claim Verification Flow.md
- Context Compaction Routine.md
- Coupling-Aware Decomposition.md
- Explore Plan Code Commit.md
- Gauntlet Fit and Reference Gate.md
- Integration Regression and Smoothing.md
- Multi-Agent Fan-Out Research Flow.md
- One-Prompt Job Contract.md
- Reporting Workflow.md
- Research Refresh Workflow.md
- Source Intake Workflow.md
- Synthesis Workflow.md
- _index.md
- AI Marketing Hub domain adapters and acceptance metrics have not yet been implem.md
- Evidence Coverage Not Yet Verified.md
- No controlled equal-budget benchmark yet isolates the Gauntlet prompt from model.md
- No general method guarantees a model critic can observe dynamic feel, long video.md
- Same-model fresh-context criticism reduces shared history but does not guarantee.md
- The best production plateau rule across domains remains unverified.md
- Answer Engine Citability Policy.md
- CONVENTIONS.md
- Corpus Scope Policy.md
- dashboard.md
- Memory Governance Policy.md
- Provenance Trace Policy.md
- Start Here.md
- Tag Taxonomy.md
- Uncertainty Eval Policy.md
- _index.md
- Importer validate gauntlet job.md
- Input Schema gauntlet job.v1.schema.md
- _index.md
- How should coupling be measured before fan-out.md
- What Current Official Source Resolves The Highest Risk Claim.md
- What equal-budget baseline would show that the loop adds value beyond more infer.md
- What evidence is sufficient for dynamic artifacts such as games, video, and inte.md
- When should a retry use the same critic, a fresh critic, multiple critics, or a.md
- Which dimensions require deterministic tests, model judges, domain experts, or r.md
- Which reference set best represents the intended audience and task rather than m.md
- _index.md
- Weekly Report.md
- _index.md
- Gauntlet Loop Evidence Baseline.md
- sample-source.md
- Source Manifest Guide.md
- hot.md
- hot.template.md
- index.md
- log.md
- log.template.md
- overview.md
- CODEX.md
- README.md
- shipping-rules.md
- weekly-report.html
- __init__.py
- __main__.py
- cli.py
- clock.py
- contract.py
- integrity.py
- 001-how-to-run-a-gauntlet-loop.md
- 002-claude-of-duty-readme-and-honest-assessment.md
- 003-claude-of-duty-original-prompt.md
- 004-building-effective-agents.md
- 005-effective-harnesses-for-long-running-agents.md
- 006-demystifying-evals-for-ai-agents.md
- 007-how-we-built-our-multi-agent-research-system.md
- 008-self-refine-iterative-refinement-with-self-feedback.md
- 009-reflexion-language-agents-with-verbal-reinforcement-learning.md
- 010-judging-llm-as-a-judge-with-mt-bench-and-chatbot-arena.md
- 011-self-in-correct-llms-struggle-with-discriminating-self-generated.md
- 012-self-preference-bias-in-llm-as-a-judge.md
- 013-specification-gaming-the-flip-side-of-ai-ingenuity.md
- 014-our-framework-for-developing-safe-and-trustworthy-agents.md
- 015-towards-a-science-of-scaling-agent-systems-when-and-why-agent-sy.md
- _index.md
- adapter-manifest.json
- adapter-plan.md
- claim-ledger.md
- CONFIDENCE_TAGS.md
- current-requirements.md
- market-research.md
- product-spec.md
- publication-readiness.json
- research-plan.md
- safety-gates.md
- source-ledger.json
- source-manifest-template.md
- source-map.md
- gauntlet-job.v1.schema.json
- audit_brain.py
- build_demo_vault.py
- gemini_loader.py
- generate_vault_visuals.py
- guide_next_action.py
- hash_sample_vault.py
- ingest_source.py
- install_manifest.py
- lint_vault.py
- package_release.py
- render_brain_report.py
- scaffold_vault.py
- synthesize_brain.py
- validate_gauntlet_job.py
- sanitize-public.mjs
- .gitignore
- package.json
- quartz.config.ts
- README.md
- gauntlet-loop-brain.yaml
- gauntlet-job.json
- sample-source.md
- sample-vault.sha256
- test_gauntlet_contract.py
- test_gemini_loader.py
- test_integrity.py
- test_pipeline.py
- test_release_semantics.py
- .gitignore
- AGENTS.md
- CHANGELOG.md
- CLAUDE.md
- CONTRIBUTING.md
- GEMINI.md
- install.sh
- LICENSE
- NOTICE
- PUBLISHING_NOTICE.md
- pyproject.toml
- README.md
- RELEASE_CHECKLIST.md
- RELEASE_NOTES.md
- SECURITY.md
- SKILL.md
- SUPPORT.md
- THIRD_PARTY_NOTICES.md
- uninstall.sh
# CDN으로 사용하기
jsDelivrjsDelivr는 공개 GitHub 리포지토리를 별도 설정 없이 CDN으로 즉시 서빙합니다. 버전과 파일을 고르면 웹페이지에 바로 붙일 수 있는 링크와 예시 코드가 만들어집니다.
링크
예시
// repository documentation
Was this content helpful?
(0 ratings)