docs: prepare for source publication

- blueprint audience generalized to 'the AI implementation agent';
  internal codename removed
- docs/README explains the untracked EXECUTION/ orchestration workspace
  so blueprint references resolve for public readers
- README states the development process itself: AI implementers under
  my lead and supervision, sealed blueprints, independent review,
  evidence-bound acceptance gates

Author: sviat-barbutsa

README.md

@@ -11,6 +11,13 @@ The product keeps project knowledge, AI session history, context bundles, diagra
 > [Frontend V2 status](docs/FRONTEND_V2_IMPLEMENTATION_STATUS.md), the
 > [developer preview boundary](docs/DEVELOPER_PREVIEW.md), and the
 > [qualification matrix](docs/qualification/frontend-qualification-matrix.md).
+>
+> The development process is part of what this repository demonstrates: the
+> frontend was built by AI implementers under my lead and supervision, working
+> from sealed implementation blueprints with independent review and
+> evidence-bound acceptance gates (see
+> [the blueprint](docs/FRONTEND_IMPLEMENTATION_BLUEPRINT.md) and the
+> qualification matrix above).
 
 Frontend V2 is an internal refactor, not a breaking release. Operation names and
 project data stay compatible; dated adapters and aliases keep existing clients,

docs/FRONTEND_IMPLEMENTATION_BLUEPRINT.md

@@ -1,6 +1,6 @@
 # Zharwing Memory Frontend Implementation Blueprint
 
-**Audience:** Codex Terra acting as implementer, reviewer, or orchestrator
+**Audience:** the AI implementation agent acting as implementer, reviewer, or orchestrator
 
 **Scope:** the React browser application, Tauri desktop shell, public website and
 documentation site, shared frontend contracts, and the minimum daemon changes
@@ -20,7 +20,7 @@ or tests.
 ## 1. Purpose
 
 This document turns the frontend audit into a complete implementation program.
-It is intended to let a fresh Codex Terra task understand the product, the
+It is intended to let a fresh implementation-agent task understand the product, the
 reasoning, the target architecture, the order of work, the evidence required,
 and the boundaries that must not be crossed without reconstructing the prior
 review.
@@ -1474,7 +1474,7 @@ missing build gate.
 
 ---
 
-## 19. Codex Terra operating contract
+## 19. Implementing-agent operating contract
 
 ### 19.1 Start and continuous campaign loop
 

docs/README.md

@@ -120,6 +120,16 @@ This folder contains product and engineering documentation for the Zharwing Memo
 10. Add durable boundary decisions to the
     [Architecture Decision Records](decisions/README.md).
 
+## About `EXECUTION/` References
+
+Some planning documents (the implementation blueprints and their baselines)
+reference files under `EXECUTION/`. That directory is the local orchestration
+workspace - work-package queues and implementation state used to drive AI
+implementation agents during development. It is deliberately not shipped in
+this repository: its contents are session-scoped working state, not product
+documentation. The blueprints remain readable without it; treat those
+references as pointers into the development process, not into this tree.
+
 ## Documentation Policy
 
 - Diagrams are Mermaid-first so they remain editable Markdown.

website/memory/docs/project-overview/index.html

@@ -24,7 +24,7 @@
       <aside class="docs-sidebar" aria-label="Documentation"><div class="docs-sidebar-inner"><a class="docs-home-link" href="/memory/docs/">Documentation home</a><div class="docs-search" role="search"><label for="docs-search-input">Search documentation</label><input id="docs-search-input" type="search" autocomplete="off" data-docs-search /><p data-search-status aria-live="polite"></p><div class="docs-search-results" data-search-results></div></div><nav class="docs-nav" aria-label="All guides"><section class="docs-nav-group"><h2>Start here</h2><a href="/memory/docs/project-overview/" aria-current="page" class="active" data-search-title="Project overview What Memory does, how it is organized, and the fastest way to start.">Project overview</a><a href="/memory/docs/setup/" data-search-title="Setup Choose a supported local profile and start the daemon, browser UI, desktop app, or agent connection.">Setup</a><a href="/memory/docs/frontend-v2-status/" data-search-title="Frontend V2 status Separate implemented source boundaries from candidate and release/device qualification.">Frontend V2 status</a><a href="/memory/docs/developer-preview/" data-search-title="Developer preview What is ready today, current limitations, and the project boundary.">Developer preview</a><a href="/memory/docs/mvp-walkthrough/" data-search-title="First walkthrough Create a project, start a session, save context, and resume work.">First walkthrough</a><a href="/memory/docs/browser-ui/" data-search-title="Browser UI Run the frequently used local interface in a browser.">Browser UI</a><a href="/memory/docs/desktop-ui/" data-search-title="Desktop UI Use the native application, navigation, and folder selection workflows.">Desktop UI</a></section><section class="docs-nav-group"><h2>Use Memory</h2><a href="/memory/docs/user-flows/" data-search-title="User flows Understand setup, daily sessions, context, review, and recovery flows.">User flows</a><a href="/memory/docs/repositories/" data-search-title="Repositories Connect one or several code repositories to a Memory project.">Repositories</a><a href="/memory/docs/workstreams/" data-search-title="Workstreams Group sessions and knowledge around long-running areas of work.">Workstreams</a><a href="/memory/docs/importing/" data-search-title="Import existing memory Preview and import existing Markdown documents and session history.">Import existing memory</a><a href="/memory/docs/graph-rules/" data-search-title="Graph rules Shape useful graph hubs with deterministic project rules.">Graph rules</a><a href="/memory/docs/semantic-graph/" data-search-title="Semantic graph Add locally reviewed semantic relationships to the context graph.">Semantic graph</a></section><section class="docs-nav-group"><h2>Connect AI agents</h2><a href="/memory/docs/mcp-setup/" data-search-title="MCP setup Connect Codex, Claude, or another MCP-compatible client.">MCP setup</a><a href="/memory/docs/agent-automation/" data-search-title="Agent automation Bootstrap repositories and make session memory part of agent work.">Agent automation</a><a href="/memory/docs/agent-protocol/" data-search-title="Agent protocol Follow the shared startup, checkpoint, and closeout contract.">Agent protocol</a><a href="/memory/docs/generic-mcp-adapter/" data-search-title="Generic MCP adapter Use the project-neutral MCP integration pattern.">Generic MCP adapter</a><a href="/memory/docs/cli-adapter/" data-search-title="CLI adapter Use Memory through command-line workflows.">CLI adapter</a><a href="/memory/docs/json-rpc-adapter/" data-search-title="JSON-RPC adapter Integrate directly with the local daemon API.">JSON-RPC adapter</a></section><section class="docs-nav-group"><h2>Architecture and decisions</h2><a href="/memory/docs/architecture/" data-search-title="Architecture See the local daemon, adapters, packages, and trust boundaries.">Architecture</a><a href="/memory/docs/data-model/" data-search-title="Data model Understand projects, sessions, documents, graph data, and storage.">Data model</a><a href="/memory/docs/api-reference/" data-search-title="API reference Reference the local HTTP, JSON-RPC, and MCP surfaces.">API reference</a><a href="/memory/docs/diagrams/" data-search-title="System diagrams Review the main component, data, and interaction diagrams.">System diagrams</a><a href="/memory/docs/source-context/" data-search-title="Source and context boundary Understand which sources become public documentation and how private context stays excluded.">Source and context boundary</a><a href="/memory/docs/architecture-decisions/" data-search-title="Architecture decisions Read the decision log and the rules for changing durable system boundaries.">Architecture decisions</a><a href="/memory/docs/adr-local-first/" data-search-title="ADR 0001: Local-first project boundary Keep application source, private memory, and public documentation in separate ownership domains.">ADR 0001: Local-first project boundary</a><a href="/memory/docs/adr-progressive-docs/" data-search-title="ADR 0002: Progressive public documentation Generate directly addressable pages whose content and navigation do not depend on JavaScript.">ADR 0002: Progressive public documentation</a><a href="/memory/docs/adr-agent-authority/" data-search-title="ADR 0003: Project-bound agent authority Use explicit project-bound agent authority and centralized privacy projection.">ADR 0003: Project-bound agent authority</a></section><section class="docs-nav-group"><h2>Security and contracts</h2><a href="/memory/docs/security/" data-search-title="Security Understand the local trust model and report vulnerabilities.">Security</a><a href="/memory/docs/principal-model/" data-search-title="Principal model Review audiences, project binding, expiry, revocation, and operation authority.">Principal model</a><a href="/memory/docs/browser-session-security/" data-search-title="Browser session security Review local browser bootstrap, CSRF, rotation, revocation, and project binding.">Browser session security</a><a href="/memory/docs/native-desktop-security/" data-search-title="Native desktop authority Review the Rust-owned daemon and credential boundary outside the webview.">Native desktop authority</a><a href="/memory/docs/provider-security/" data-search-title="Provider secrets and egress Operate write-only credentials and constrained provider destinations.">Provider secrets and egress</a><a href="/memory/docs/privacy-projection/" data-search-title="Privacy projection Understand centralized agent and provider projection, visibility, and completeness.">Privacy projection</a><a href="/memory/docs/operation-admission/" data-search-title="Operation admission Trace requests through exact decoding, authority, idempotency, and dispatch.">Operation admission</a></section><section class="docs-nav-group"><h2>Operate and contribute</h2><a href="/memory/docs/operations/" data-search-title="Operations Run, package, back up, restore, and troubleshoot Memory.">Operations</a><a href="/memory/docs/ai-testing/" data-search-title="AI provider testing Test local and external model-backed workflows safely.">AI provider testing</a><a href="/memory/docs/testing-plan/" data-search-title="Testing plan See the testing strategy and expected coverage.">Testing plan</a><a href="/memory/docs/frontend-accessibility/" data-search-title="Frontend accessibility Review the responsive, keyboard, focus, contrast, motion, and manual qualification contract.">Frontend accessibility</a><a href="/memory/docs/frontend-v2-migration/" data-search-title="Frontend V2 migration Migrate profiles, credentials, visibility, preferences, and layout caches with rollback.">Frontend V2 migration</a><a href="/memory/docs/frontend-v2-compatibility/" data-search-title="Compatibility register See retained and removed compatibility paths, callers, dates, and removal proof.">Compatibility register</a><a href="/memory/docs/frontend-qualification/" data-search-title="Frontend qualification Separate source observations from candidate, browser, WebView, assistive-device, and release evidence.">Frontend qualification</a><a href="/memory/docs/contributing/" data-search-title="Contributing Set up the repository and follow the contribution workflow.">Contributing</a></section></nav></div></aside>
       <main class="docs-main" id="docs-content"><article class="doc-article"><header class="article-header"><p class="article-kicker">Start here</p><h1>Project overview</h1><p class="article-lede">What Memory does, how it is organized, and the fastest way to start.</p><div class="article-meta"><span>11 min read</span><a href="https://github.com/zharwing/memory/blob/main/README.md" target="_blank" rel="noreferrer">View source on GitHub ↗</a></div></header><details class="mobile-toc"><summary>On this page</summary><nav><a class="toc-level-2" href="#current-implementation-status">Current Implementation Status</a><a class="toc-level-3" href="#what-has-actually-been-tested">What has actually been tested</a><a class="toc-level-2" href="#product-principles">Product Principles</a><a class="toc-level-2" href="#repository-layout">Repository Layout</a><a class="toc-level-2" href="#first-run">First Run</a><a class="toc-level-3" href="#local-browser-ui">Local Browser UI</a><a class="toc-level-3" href="#native-desktop-ui">Native Desktop UI</a><a class="toc-level-3" href="#pointer-files">Pointer Files</a><a class="toc-level-2" href="#architecture-summary">Architecture Summary</a><a class="toc-level-2" href="#memory-root-shape">Memory Root Shape</a><a class="toc-level-2" href="#main-workflow">Main Workflow</a><a class="toc-level-2" href="#agent-automation">Agent Automation</a><a class="toc-level-2" href="#cli-examples">CLI Examples</a><a class="toc-level-2" href="#mcp-tools">MCP Tools</a><a class="toc-level-2" href="#browser-and-desktop-ui">Browser And Desktop UI</a><a class="toc-level-2" href="#documentation">Documentation</a><a class="toc-level-2" href="#implementation-notes">Implementation Notes</a><a class="toc-level-2" href="#license">License</a></nav></details><div class="markdown-body"><p>Zharwing Memory is a local-first project context manager for AI-assisted coding workflows. It is not the coding agent. External agents such as Codex, Claude Code, Gemini CLI, Ollama-based tools, LM Studio workflows, and future MCP-capable clients do the engineering work. Zharwing Memory provides their durable, project-scoped memory layer.</p>
 <p>The product keeps project knowledge, AI session history, context bundles, diagrams, decisions, commands, gotchas, and optional review proposals organized per project. A human can open the local browser UI or native desktop app to understand current work, inspect AI context, inspect the graph, search previous work, and enable review workflows only when wanted.</p>
-<blockquote><p><strong>Status: local developer preview. Not release-qualified.</strong> The code ships two profiles: <code>personal-preview</code> for compatibility and <code>hardened-local</code> as the security target. Both are implemented; neither has been through installer, device, or production testing. See <a href="/memory/docs/frontend-v2-status/">Frontend V2 status</a>, the <a href="/memory/docs/developer-preview/">developer preview boundary</a>, and the <a href="/memory/docs/frontend-qualification/">qualification matrix</a>.</p></blockquote>
+<blockquote><p><strong>Status: local developer preview. Not release-qualified.</strong> The code ships two profiles: <code>personal-preview</code> for compatibility and <code>hardened-local</code> as the security target. Both are implemented; neither has been through installer, device, or production testing. See <a href="/memory/docs/frontend-v2-status/">Frontend V2 status</a>, the <a href="/memory/docs/developer-preview/">developer preview boundary</a>, and the <a href="/memory/docs/frontend-qualification/">qualification matrix</a>.  The development process is part of what this repository demonstrates: the frontend was built by AI implementers under my lead and supervision, working from sealed implementation blueprints with independent review and evidence-bound acceptance gates (see <a href="https://github.com/zharwing/memory/blob/main/docs/FRONTEND_IMPLEMENTATION_BLUEPRINT.md" target="_blank" rel="noreferrer">the blueprint</a> and the qualification matrix above).</p></blockquote>
 <p>Frontend V2 is an internal refactor, not a breaking release. Operation names and project data stay compatible; dated adapters and aliases keep existing clients, bookmarks, and local preview setups working through the migration.</p>
 <figure class="doc-image"><img src="/memory/assets/dashboard.png" alt="Zharwing Memory project dashboard" width="1280" height="720" loading="lazy" decoding="async" /></figure>
 <p>The <a href="https://zharwing.barbutsa.com/memory/" target="_blank" rel="noreferrer">documentation website</a> and its <a href="https://zharwing.barbutsa.com/memory/docs/" target="_blank" rel="noreferrer">guide portal</a> are built from this repository and pull in no external dependencies. They document the local app; they are not a hosted copy of it and cannot reach your memory store. Every guide has its own URL and reads fine with JavaScript turned off. See the <a href="https://github.com/zharwing/memory/blob/main/website/memory/README.md" target="_blank" rel="noreferrer">website maintenance guide</a> and the <a href="/memory/docs/source-context/">source-context boundary</a>.</p>