06 / LIBRARY
The reading room.
Every recommendation on this site traces back to a document in here. Plans are the decisions — synthesis, architecture, pre-mortems. Research is the raw evidence those decisions were built on. Nothing above this floor is invented.
Plan folders
7
Plan documents
85
Research documents
16
Plans
strategy & decisions · 7 folders
001
Factory Apps Validation
6 docs
- Factory apps validation — architecture decision001-factory-apps-validation/readme
- Factory apps validation — Synthesis & Decision001-factory-apps-validation/00-synthesis
- Deep dive — codebase & ground truth001-factory-apps-validation/01-deep-dive
- Case studies & experience reports001-factory-apps-validation/02-case-studies
- Options — candidate portfolios001-factory-apps-validation/03-options
- Steelman court001-factory-apps-validation/04-steelman
002
Dual Run Parity Harness
17 docs
- Dual-run parity harness — product research ("the quality-control bench")002-dual-run-parity-harness/readme
- Context brief — Dual-run parity harness product research (plans/002)002-dual-run-parity-harness/00-context-brief
- Product architecture — Dual-run parity harness ("quality-control bench")002-dual-run-parity-harness/architecture
- Build vs buy — dual-run parity harness002-dual-run-parity-harness/build-vs-buy
- Competitive analysis — dual-run parity harness002-dual-run-parity-harness/competitive-analysis
- Executive recommendation — Dual-run parity harness002-dual-run-parity-harness/executive-recommendation
- Initial recommendation — Dual-run parity harness002-dual-run-parity-harness/initial-recommendation
- Journey — Dual-run parity harness002-dual-run-parity-harness/journey
- Made to Stick — Dual-run parity harness002-dual-run-parity-harness/made-to-stick
- North Star — Dual-run parity harness002-dual-run-parity-harness/north-star
- Open-source foundation analysis — dual-run parity harness002-dual-run-parity-harness/open-source-analysis
- PRD — Dual-run parity harness ("the quality-control bench")002-dual-run-parity-harness/prd
- Pre-mortem — Dual-run parity harness002-dual-run-parity-harness/pre-mortem
- Research — Dual-run parity harness (the intellectual core)002-dual-run-parity-harness/research
- User stories — dual-run parity harness002-dual-run-parity-harness/user-stories
- UX/UI specification — dual-run parity harness002-dual-run-parity-harness/ux-ui
- Validation plan — dual-run parity harness002-dual-run-parity-harness/validation-plan
002
Work Ledger Product
16 docs
- Work-ledger — product research & go/no-go002-work-ledger-product/readme
- Work-ledger — Executive recommendation (Synthesis)002-work-ledger-product/00-synthesis
- Initial recommendation — pre-research expert take002-work-ledger-product/01-initial-recommendation
- Research — brainstorm, problem discovery, evidence tiers, advisory board, steelman002-work-ledger-product/02-research
- Stakeholders & journey — work-ledger adoption inside the two-lead fleet002-work-ledger-product/03-journey
- User stories002-work-ledger-product/04-user-stories
- Competitive & alternative-solution analysis002-work-ledger-product/05-competitive-analysis
- 06 — Open-source foundation analysis002-work-ledger-product/06-open-source-analysis
- UX / UI — work-ledger002-work-ledger-product/07-ux-ui
- Made to stick — work-ledger SUCCESs pass002-work-ledger-product/08-made-to-stick
- North star — strategy, positioning, North Star Metric002-work-ledger-product/09-north-star
- 10 — Product architecture002-work-ledger-product/10-architecture
- 11 — Build vs buy002-work-ledger-product/11-build-vs-buy
- 12 — PRD: work-ledger002-work-ledger-product/12-prd
- Pre-mortem — work-ledger, 18 months on002-work-ledger-product/13-pre-mortem
- 14 — Validation plan: riskiest assumptions → cheapest disproving tests002-work-ledger-product/14-validation-plan
003
Atlas Intelligence Layer
24 docs
- Atlas — intelligence & communication layer: product research & go/no-go003-atlas-intelligence-layer/readme
- 00 — SYNTHESIS: Executive Recommendation on Atlas003-atlas-intelligence-layer/00-synthesis
- 01 — Initial Product & UX Recommendation: Atlas003-atlas-intelligence-layer/01-initial-recommendation
- 02 - Research: Problem Discovery, Advisory Board, Steelman003-atlas-intelligence-layer/02-research
- Stakeholders & journey: Atlas003-atlas-intelligence-layer/03-journey
- Atlas User Stories003-atlas-intelligence-layer/04-user-stories
- 05 - Competitive and Alternative-Solution Analysis: Atlas003-atlas-intelligence-layer/05-competitive-analysis
- 06 - Open-Source Foundation Analysis003-atlas-intelligence-layer/06-open-source-analysis
- 07 — UX/UI Specification: Atlas003-atlas-intelligence-layer/07-ux-ui
- 08 - Made to Stick: Atlas SUCCESs Analysis003-atlas-intelligence-layer/08-made-to-stick
- 09 -- Atlas: Product Strategy and North Star003-atlas-intelligence-layer/09-north-star
- 10 - Atlas Architecture: the answer loop on the Zaruba substrate003-atlas-intelligence-layer/10-architecture
- 11 — Build vs Buy: the Atlas Answer Loop003-atlas-intelligence-layer/11-build-vs-buy
- 12 - Product Requirements: Atlas (the answer loop)003-atlas-intelligence-layer/12-prd
- 13 - Pre-mortem: the Atlas inquest (January 2028)003-atlas-intelligence-layer/13-pre-mortem
- 14 - Validation Plan: Atlas (the answer loop)003-atlas-intelligence-layer/14-validation-plan
- Atlas evidence: substrate investigation (r1)003-atlas-intelligence-layer/evidence/r1-substrate
- R2 — Prior internal decisions & evidence relevant to Atlas003-atlas-intelligence-layer/evidence/r2-prior-decisions
- R3 — Human-in-the-loop question triage for AI agent fleets: 2025-2026 landscape003-atlas-intelligence-layer/evidence/r3-hitl-landscape
- R4 — AI Codebase Documentation Tools & Docs-for-Agents Conventions (2025-2026)003-atlas-intelligence-layer/evidence/r4-ai-docs
- R5 — Team status / progress-communication products: evidence and why they die003-atlas-intelligence-layer/evidence/r5-comms-products
- R6 — Knowledge bases, Q&A products, glossary practice (2025-2026)003-atlas-intelligence-layer/evidence/r6-knowledge-qa
- R7 — OSS foundation candidates for "agent question loop + knowledge write-back"003-atlas-intelligence-layer/evidence/r7-oss
- R8 — Negative Evidence File: The Case Against Building Atlas003-atlas-intelligence-layer/evidence/r8-negative
004
Fleet Monitor
9 docs
- 004 — Fleet Monitor (factory app #4)004-fleet-monitor/readme
- 00 — SYNTHESIS: Fleet Monitor — Executive Recommendation004-fleet-monitor/00-synthesis
- 01 — Initial Product & UX Recommendation: Fleet Monitor004-fleet-monitor/01-initial-recommendation
- 02 — Discovery: Fleet Monitor004-fleet-monitor/02-discovery
- 03 — Competitive, Alternative and Open-Source Foundation Analysis004-fleet-monitor/03-competitive-oss
- 04 — PRD: Fleet Monitor004-fleet-monitor/04-prd
- 05 — UX/UI Specification: Fleet Monitor004-fleet-monitor/05-ux-ui
- 06 — Fleet Monitor Architecture004-fleet-monitor/06-architecture
- 07 — Pre-Mortem: Fleet Monitor004-fleet-monitor/07-pre-mortem
005
Showcase Website
6 docs
- 005 — Showcase Website (stack + repo setup)005-showcase-website/readme
- 00 — SYNTHESIS: Showcase Website Stack + Repo Setup005-showcase-website/00-synthesis
- 005 — Showcase Website: Phase-1 Deep Dive005-showcase-website/01-deep-dive
- 02 — Case Studies: Agentic-Era Experience + Classic Adopter Reports005-showcase-website/02-case-studies
- Options — showcase website candidates005-showcase-website/03-options
- 04 — Steelman: Adversarial Court Record005-showcase-website/04-steelman
006
Stack Architecture
7 docs
- Stack architecture for the four factory apps — planning folder006-stack-architecture/readme
- 00 — SYNTHESIS: Stack architecture for the four factory apps006-stack-architecture/00-synthesis
- 01 — Deep Dive: Binding Constraints, Host Ground Truth, Footprint Evidence006-stack-architecture/01-deep-dive
- Case studies & agentic-era language-fit evidence006-stack-architecture/02-case-studies
- 03 — Stack Candidates006-stack-architecture/03-options
- 04 — Steelman Court006-stack-architecture/04-steelman
- 05 — Postgres-for-everything vs multi-store: comparison and ruling006-stack-architecture/05-postgres-only-comparison
Research
raw evidence log · 16 documents
Groupon Agentic Refactoring — Dossier & Factory Blueprint00-dossierExecutive Summary00-executive-summaryGround Truth — What Exists Today (2026-07-18)01-ground-truthDependency Inventory and Codebase Analysis01-inventory-dependency-analysisPlatform Architecture and Agent Orchestration02-platform-architecture-orchestrationWeb Research — Agentic Mass Refactoring, Verified Evidence (2026-07-18)02-web-researchObservability, Triage, and Safety Controls03-observability-triage-safetyModels, Tooling, and Cost Model04-models-tooling-costPhased Rollout, KPIs, and Risk Management05-rollout-kpis-risksSix-Application Control Plane (Research Batch 2)06-six-app-blueprintCodebase Knowledge Graph + Declarative Rewrite Engines07-knowledge-graph-rewrite-enginesOrchestration, Triage Gateway, and Guardrails08-orchestration-triage-guardrailsMonitoring, Rate Limits, and Comparative Metrics09-monitoring-limits-comparative-metricsSafe Execution of Interdependent Core Logic (1Password B5 Case Study)10-core-logic-case-study12 — Two-lead work split + fleet monitor verdict12-collab-split-and-fleet-monitorResearch: Agentic Large-Scale Automated Refactoring (TS + Go Monorepo)readme