// technical blog / research notes
.knowledge technical notes
Research notes, practical comparisons, real cases, benchmarks, design decisions, and release hygiene notes for teams deciding why a trust layer matters.
// site and GitHub roles
Decision layer, proof layer
The website explains the product, trust model, examples, comparisons, and design tradeoffs. GitHub keeps the install asset, source, schemas, CLI, tests, fixtures, validators, exact commands, and reproducible proof.
// all notes
Research and implementation notes
.knowledge benchmarks and measurement notes
How to read local benchmark and metrics claims without turning them into universal performance guarantees.
Graph explanationWhat the free-core graph should do
What belongs in the free/core graph and why decorative graph visuals are not enough.
Release hygieneRelease artifact hygiene
Why the install archive must be a validated release asset, not a GitHub source snapshot.
Core discoverabilityWhat is repo-local knowledge routing for coding agents?
A practical definition of repo-local knowledge routing: the first-read path, artifacts, trust rules, and where .knowledge fits in an AI coding workflow.
Core discoverabilityCodex repo onboarding checklist for .knowledge
A Codex-ready checklist for adding .knowledge to a repository, installing agent integrations, running import and release flows, and checking trust before edits.
Core discoverabilityHow to reduce repeated repo crawls by coding agents
A practical way to reduce repeated first-orientation crawls by giving coding agents one routing bundle, scoped source checks, and cautious metrics.
Core discoverabilityOpen-source agent context layer for coding teams
What an open-source, repo-local agent context layer should include: routing, trust, freshness, tools, integrations, Inspector, and clear free vs paid boundaries.
ComparisonLocal-first agent memory vs RAG for coding agents
A practical comparison of local-first agent memory and RAG for coding agents, with trust rules, artifact paths, and when each model should be used.
ComparisonClaude Code vs Codex for repo context
A practical comparison of Claude Code and Codex for repo context, and how a shared .knowledge layer keeps both agents aligned.
Inspector workflowsVisual inspector for AI coding agent trust and freshness
How a visual inspector helps teams see trust, freshness, repair work, and stale modules before agents make code changes.
GovernanceHow to audit agent repo context before code changes
A practical pre-change audit for AI coding agents: routing, trust, freshness, evidence, source files, tests, and repair queue.
Team workflowsHow to review PR impact for coding agents
A practical PR impact review workflow for AI coding agents: changed files, module boundaries, trust state, evidence, tests, and reviewer handoff.
Team workflowsAgent handoff between coding sessions without losing repo context
How to hand off coding-agent work between sessions using repo-local routing, trust reports, repair queues, and source-of-truth rules.
MaintenanceHow to keep agent repo context fresh after commits and merges
A maintenance workflow for keeping AI coding agent context fresh after commits, merges, module changes, and documentation updates.
FAQFAQ: .knowledge for teams
A practical FAQ for teams evaluating .knowledge: routing, trust, freshness, open core, Inspector, RAG, agent handoff, and PR review.