soroush.asadi fe7a5c481e M6: working memory + the PO→QA trigger + analytics — V1 complete
Working memory (Memory module's first real code):
- MemoryEntry (schema "memory", vector(384), InitialMemory migration); TeamMemory implements
  the SharedKernel ITeamMemory seam (embed-and-store on write, cosine recall on read);
  GET /api/memory/search. HashingTextEmbedder promoted to SharedKernel (pure, deterministic;
  swapped for ONNX/BYOK embedders later behind ITextEmbedder).
- Written on approval: Governance's approve stores an Approval/Correction entry per decision.
- Read at assembly: the executor recalls the team's top-3 relevant entries; the prompt gains
  a "# Team memory" section (treated as data, not instructions).

The single V1 event trigger:
- IAgentDispatcher (SharedKernel) implemented by Assembler's AgentRunDispatcher (shared by
  the API and triggers). OrgBoard's QaHandoffTrigger: a task hitting done creates a QA task
  (provenance parent, assigned to the QA agent) and dispatches a run for the team's QA AI
  seat. Guardrails: Test/Review tasks never re-trigger (no self-cascade) and a task hands
  off at most once. Audited as handoff.triggered.

Analytics — the V1 verdict view:
- IBoardStats (SharedKernel) implemented by OrgBoard; GET /api/governance/analytics returns
  approval rate, avg edit distance, per-agent metrics + edit-distance trend, tasks done.
- UI: /analytics — stat cards, per-agent table, recharts edit-distance trend per agent.

Verified: build green; ArchitectureTests 8/8; IntegrationTests 42/42 incl. the M6 acceptance
end to end — a dev marks a story done → Quill wakes via the handoff (QA task with provenance,
assigned to the agent) → drafts a test plan that waits in review → approve records the second
agent's edit distance → analytics show approval rate 100%, avg edit distance > 0, and trends
for BOTH Aria and Quill; memory written on Aria's corrected approval is recalled into her next
prompt; the guardrails hold. Client build green.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-10 12:07:35 +03:30
2026-06-09 06:41:28 +03:30
2026-06-09 06:41:28 +03:30
2026-06-09 06:41:28 +03:30

TeamUp.AI

Build human + AI teams. A live org chart that does work: model the org, fill open role-seats with governed AI agents, run delivery on one board. A product of AliaSaaS.

Status: pre-M1 skeleton — the repo builds, tests green, and runs, but carries no feature code yet. See docs/V1_BUILD_PLAN.md for what M1M6 add.

Stack

.NET 10 modular monolith (web + worker on one image) · PostgreSQL 17 + pgvector · EF Core 10 · React 19 + Vite SPA into wwwroot · Docker Compose for local dev. Full bill of materials in docs/V1_BUILD_PLAN.md.

Layout

src/Shared/TeamUp.SharedKernel     IModule seam, base Entity, IModuleDbContext
src/Shared/TeamUp.Infrastructure   PlatformDbContext (pgvector + schemas), MigrationRunner, wiring
src/Bootstrap/TeamUp.Bootstrap     the explicit module catalog (the only thing that knows all modules)
src/Modules/TeamUp.Modules.*       OrgBoard · Identity · Skills · Assembler · Governance · Memory · Integrations
src/Hosts/TeamUp.Web               ASP.NET Core API host (also serves the SPA)
src/Hosts/TeamUp.Worker            Generic Host worker (background jobs; M4+)
client/                            React/Vite SPA → builds into TeamUp.Web/wwwroot
tests/                             ArchitectureTests (boundary rules) · IntegrationTests (Testcontainers)
docker/                            Dockerfile (one image, two roles) · docker-compose.yml

Boundary rule: each module is its own assembly; everything but its IModule and public contracts is internal, so no module can touch another's persistence. TeamUp.ArchitectureTests backstops this — it fails the build if a module references another module / Infrastructure / a host.

Prerequisites

.NET SDK 10 · Node 22 · Docker. Packages and container images are pulled through a Nexus mirror — see Package & image sources below.

Quick start

# Backend: build, run the boundary + integration tests (integration needs Docker)
dotnet build TeamUp.slnx
dotnet test  TeamUp.slnx

# Local dev — two terminals, with the Vite dev server proxying /api to the .NET host
docker compose -f docker/docker-compose.yml up postgres -d   # Postgres 17 + pgvector
dotnet run --project src/Hosts/TeamUp.Web                     # http://localhost:5180  (applies migrations in Dev)
cd client && npm install && npm run dev                       # http://localhost:5173  (proxies /api, /health)

# Or run the whole thing in containers (web + worker + postgres, single image, RUN_MODE picks the role)
docker compose -f docker/docker-compose.yml up --build

GET /health is the DB health check; GET /api/<module>/ping proves each module seam; the OpenAPI document is at /openapi/v1.json (Development only).

EF Core migrations

dotnet ef migrations add <Name> \
  --project src/Shared/TeamUp.Infrastructure --startup-project src/Shared/TeamUp.Infrastructure \
  --context PlatformDbContext --output-dir Persistence/Migrations

The initial InitialPlatform migration enables the vector extension and creates one schema per module. Module-owned contexts (M1+) get their own migrations and apply after Platform via MigrationRunner.

Package & image sources

NuGet packages (nuget.config) and container base images (docker/Dockerfile, docker/docker-compose.yml) are pulled through the self-hosted Nexus mirror mirror.soroushasadi.com, which proxies nuget.org / Docker Hub / MCR and is reachable from Iran. To build against the public registries instead, point nuget.config at api.nuget.org and replace the mirror.soroushasadi.com/... image prefixes with docker.io/library (node) and mcr.microsoft.com (dotnet). Testcontainers picks up the mirror via TESTCONTAINERS_HUB_IMAGE_NAME_PREFIX=mirror.soroushasadi.com/.

More

S
Description
No description provided
Readme 696 KiB
Languages
C# 67%
TypeScript 29.8%
CSS 2.6%
Dockerfile 0.3%
HTML 0.2%