git:20260824.6d915b2 to git:20260916.e7b3bfe

17 added, 19 removed. Audit A to A.

- # AIDE Sovereign Workbench

+ # Covert Coder — AIDE Sovereign Workbench

- > Offline-first, open-source development workbench: real code editor, local AI models, Git review, debugging, and optional online providers — without sending source code to the cloud.

+ > Covert Coder is the product direction for AIDE Sovereign Workbench, an open-source local-first development environment with editor, models, Git, debugging, and verification under operator control.

- ## Key facts

- - Two-process architecture: browser frontend (Monaco editor) + Node.js daemon on loopback (HTTP/WebSocket, ports 4173/4777).

- - Cockpit orchestrator UI: Describe -> Plan -> Approve -> Build -> Verify; approval-gated agent loop with diff previews, live git-diff workbench, verification rail.

- - Model Hub in-app: Hugging Face GGUF search, quantization listing, resumable downloads, one-click engine registration; local import by path.

- - Harness scaffold v2.1: sized operating layer per model (micro for small contexts, full Code+Lens for strong budgets); byte-deterministic; 20-task effectiveness battery published in docs/evidence/.

- - Backend auto-select: probes llama-server builds via --list-devices; Vulkan/CUDA/CPU chosen by operator profile > benchmark verdict > hardware heuristic.

- - Local AI chat via llama.cpp OpenAI-compatible loopback; GGUF weights downloaded separately and SHA-256 verified; never bundled.

- - LSP client for TypeScript; DAP client for Python/debugpy. Git via CLI. Tasks service with problem matchers.

- - Online providers opt-in (OpenAI, Anthropic, Gemini, Mistral, Groq, OpenRouter); DPAPI-encrypted credentials; host allowlist.

- - Veritas evidence gates: models never approve or apply their own changes.

- - Verified continuously: 188 architecture tests, 17 e2e browser tests, 10/10 doctor preflight, plus unit batteries (facade 12, scaffold 5, agent loop 14+5).

+ - Current status: pre-production; capability labels are conservative and uncommitted work is not release evidence.

+ - Product topology: browser frontend (:4173) → product façade (:4777) → canonical TypeScript backend (:4778); legacy backend (:4779) remains migration inventory.

+ - Local model execution and local tools are supported; external model-host downloads and online providers are explicit opt-in network boundaries.

+ - Resident is advisory; models, plugins, and generated patches do not grant authority or apply changes without operator review.

+ - Implemented route families cover workspace, editor, chat, models, providers, BYOK, Git, terminal, tasks, LSP, DAP, Resident, memory, audit, and verification-related services.

+ - Veritas/Harness provides evidence gates; verification is not execution permission or effect certainty.

+ - Desktop packaging, Helix extensions, dedicated Skills/Security surfaces, and the Covert cockpit remain experimental or unreleased where the feature matrix says so.

+ - No test count is asserted here; run the documented checks and retain their evidence for release claims.

## Links

- - Repository: https://github.com/AnonymousNomad/aide-sovereign-workbench

- - License: Apache-2.0 (LICENSE)

+ - Repository: https://github.com/AnonymousNomad/covert-coder

+ - Documentation index: docs/README.md

+ - Architecture: docs/ARCHITECTURE.md

- Security: SECURITY.md

- Contributing: CONTRIBUTING.md

- - Roadmap: docs/RELEASE_ROADMAP.md

- - Research log: docs/RESEARCH_LOG.md

+ - Release roadmap: docs/RELEASE_ROADMAP.md

- Model packs: models/PACKS.md

- - Benchmarks: benchmarks/

- - Docs: docs/

+ - License: LICENSE