OSS contributors who clone molecule-core and `go run ./workspace-server/cmd/server`
now get a working end-to-end provision without authenticating to GHCR or AWS ECR.
Pre-fix: with MOLECULE_IMAGE_REGISTRY unset, the provisioner attempted to pull
ghcr.io/molecule-ai/workspace-template-<runtime>:latest, which has been
returning 403 since the 2026-05-06 GitHub-org suspension.
Post-fix: when MOLECULE_IMAGE_REGISTRY is unset, the provisioner switches to
local-build mode — looks up the workspace-template-<runtime> repo's HEAD sha
on Gitea via a single API call, shallow-clones into ~/.cache/molecule/, and
runs `docker build --platform=linux/amd64`. SHA-pinned cache key skips the
clone+build entirely on subsequent provisions.
Production tenants are unaffected: every prod tenant sets the var to its
private ECR mirror, so the SaaS pull path is byte-for-byte identical.
SSOT for mode detection lives in Resolve() (registry_mode.go) returning a
discriminated RegistrySource{Mode, Prefix} so call sites that branch on
mode get a compile-time push instead of a string-equality footgun.
Coverage:
* registry_mode.go — new SSOT (Resolve, RegistryMode, IsKnownRuntime)
* registry_mode_test.go — 8 tests pinning mode-decision contract
* localbuild.go — clone+build pipeline (570 LOC, fully unit-tested)
* localbuild_test.go — 22 tests covering happy/sad paths, fail-closed
* provisioner.go — Start() inserts ensureLocalImageHook in local mode
* docs/adr/ADR-002 — design rationale + alternatives + security review
* docs/development/local-development.md — local-build flow + env overrides
Security:
* Allowlist-only runtime names (knownRuntimes) gate the clone path.
* Repo prefix hardcoded to git.moleculesai.app/molecule-ai/molecule-ai-workspace-template-;
forks via opt-in MOLECULE_LOCAL_TEMPLATE_REPO_PREFIX.
* MOLECULE_GITEA_TOKEN masked in every log line via maskTokenInURL/maskTokenInString.
* Fail-closed: Gitea unreachable / runtime not mirrored → clear error, never
silently fall back to GHCR/ECR.
* docker build invocation passes no --build-arg from external input.
* HTTP body cap 64KB on Gitea API responses (defence vs malicious upstream).
Closes #63 / Task #194.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
7.8 KiB
Local Development
Workspace Template Images: Local-Build Mode (Issue #63)
OSS contributors who run molecule-core locally do not need to authenticate to GHCR or AWS ECR. When the MOLECULE_IMAGE_REGISTRY env var is unset, the platform automatically:
- Looks up the HEAD sha of
https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-template-<runtime>(single API call, no clone). - If a local image tagged
molecule-local/workspace-template-<runtime>:<sha12>already exists, reuses it (cache hit). - Otherwise, shallow-clones the repo into
~/.cache/molecule/workspace-template-build/<runtime>/<sha12>/and runsdocker build --platform=linux/amd64 -t <tag> .. - Hands the SHA-pinned tag to Docker for
ContainerCreate.
First-provision build time: 5–10 min on Apple Silicon (amd64 emulation). Subsequent provisions hit the cache and start in seconds. Cache is invalidated automatically when the template repo's HEAD moves.
Currently mirrored on Gitea: claude-code, hermes, langgraph, autogen. Other runtimes (crewai, deepagents, codex, gemini-cli, openclaw) fail with an actionable "not mirrored to Gitea" error pointing at the missing repo.
Production tenants are unaffected — every prod tenant sets MOLECULE_IMAGE_REGISTRY to its private ECR mirror via Railway env / EC2 user-data, so the SaaS pull path stays identical.
Environment overrides
| Var | Default | Use case |
|---|---|---|
MOLECULE_IMAGE_REGISTRY |
(unset) | Set to a real registry URL to switch from local-build to SaaS-pull mode. |
MOLECULE_LOCAL_BUILD_CACHE |
~/.cache/molecule/workspace-template-build |
Override cache directory. |
MOLECULE_LOCAL_TEMPLATE_REPO_PREFIX |
https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-template- |
Point at a fork. |
MOLECULE_GITEA_TOKEN |
(unset) | Required only if your fork has private template repos. |
Verifying a switch from the GHCR-retag stopgap
Pre-fix, OSS contributors worked around the suspended GHCR org by manually retagging an :latest image. After this change, that workaround is redundant: simply unset MOLECULE_IMAGE_REGISTRY (or leave it unset), boot the platform, and provision a workspace. Logs will show:
Provisioner: local-build mode → using locally-built image molecule-local/workspace-template-claude-code:<sha12> for runtime claude-code
local-build: cloning https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-template-claude-code → ...
local-build: docker build done in <duration>
If you still see ghcr.io/molecule-ai/... in the boot log, double-check env | grep MOLECULE_IMAGE_REGISTRY — a stale shell export from the pre-fix workaround could keep SaaS-mode active.
Starting the Stack
docker compose up
This starts:
| Service | Port | Description |
|---|---|---|
| Postgres | internal only | Primary database |
| Redis | internal only | Ephemeral state |
| Platform (Go) | :8080 |
Control plane API |
| Canvas (Next.js) | :3000 |
Visual frontend |
| Langfuse web | :3001 (host) / :3000 (internal) |
Observability UI |
| Langfuse worker | — | Background processing |
| ClickHouse | — | Langfuse dependency |
Each workspace container is provisioned on demand by the platform when a user creates or imports one.
Langfuse uses a dedicated langfuse Postgres database. The compose stack creates it automatically before starting the Langfuse service, so it does not conflict with the platform's molecule schema.
Infrastructure Only
To start just Postgres, Redis, and Langfuse (no application code):
docker compose -f docker-compose.infra.yml up
Optional Profiles
docker compose --profile multi-provider up # Add LiteLLM proxy (unified LLM API)
docker compose --profile local-models up # Add Ollama (local LLM models)
Environment Variables
Platform (Go)
DATABASE_URL=postgres://dev:dev@postgres:5432/molecule?sslmode=prefer
REDIS_URL=redis://redis:6379
PORT=8080
SECRETS_ENCRYPTION_KEY=dev-key-change-in-production
WORKSPACE_DIR=/path/to/molecule-monorepo # Optional global fallback; prefer per-workspace workspace_dir in org.yaml or API
Canvas (Next.js)
NEXT_PUBLIC_PLATFORM_URL=http://localhost:8080
NEXT_PUBLIC_WS_URL=ws://localhost:8080/ws
Workspace Runtime
WORKSPACE_ID= # assigned by platform on provision
WORKSPACE_CONFIG_PATH= # path to config folder inside container
MODEL_PROVIDER= # e.g. anthropic:claude-sonnet-4-6
TIER= # 1, 2, 3, or 4
PLATFORM_URL= # http://platform:8080
PARENT_ID= # set by platform during team expansion (empty for top-level)
ANTHROPIC_API_KEY= # or OPENAI_API_KEY, etc.
LANGFUSE_HOST= # http://langfuse-web:3000 (internal container port; host-mapped to :3001)
LANGFUSE_PUBLIC_KEY=
LANGFUSE_SECRET_KEY=
LANGSMITH_TRACING=true # LangGraph reads this to enable tracing
Technology Versions
Go 1.25+ (go.mod)
Python 3.11+
Node.js 22+
Next.js 15
React Flow 12 (@xyflow/react)
a2a-sdk 0.3+ (A2A server SDK, install with a2a-sdk[http-server])
langfuse 3.x (self-hosted Docker)
Postgres 16
Redis 7
Docker Compose 2.x
Running Tests
Unit Tests
cd workspace-server && go test -race ./... # Go tests with race detection (695 tests)
cd canvas && npm test # Vitest tests (357 tests)
cd workspace && python -m pytest -v # Workspace runtime tests (1140 tests)
cd sdk/python && python -m pytest -v # SDK tests (121 tests)
cd mcp-server && npm test # MCP server tests (97 Jest tests)
Integration Tests
bash tests/e2e/test_api.sh # 62 API tests (quick local verify; Phase 30.1 bearer-auth aware; also runs in CI)
bash tests/e2e/test_a2a_e2e.sh # 22 A2A e2e tests (requires platform + 2 agents)
bash tests/e2e/test_activity_e2e.sh # 25 activity/task E2E tests (requires platform + 1 agent)
bash tests/e2e/test_comprehensive_e2e.sh # 67 endpoint/memory/bundle/approval checks
All E2E scripts share tests/e2e/_lib.sh helpers and are shellcheck-clean (enforced in CI). See ./testing-e2e.md for auth prerequisites and what CI runs.
CI Pipeline
GitHub Actions runs automatically on push to main and on PRs (.github/workflows/ci.yml):
- platform-build — Go build, vet,
go test -racewith coverage profiling (25% baseline threshold; setup-go uses module cache) - canvas-build — npm build,
vitest run(no--passWithNoTests-- tests must exist and pass) - mcp-server-build — npm build
- python-lint —
pytest --cov=. --cov-report=term-missing(pytest-cov enabled) - e2e-api (added 2026-04-13) — Postgres + Redis service containers, migrations applied via
docker exec,tests/e2e/test_api.shmust pass 62/62 - shellcheck (added 2026-04-13) — lints every
tests/e2e/*.sh
Postgres and Redis are not exposed to the host -- use docker compose exec postgres psql or docker compose exec redis redis-cli for direct access.
Utility Scripts
| Script | Purpose |
|---|---|
infra/scripts/setup.sh |
Initialize the local environment |
infra/scripts/nuke.sh |
Tear down and clean up everything |
bundle-compile.sh |
Compile workspace config folders into .bundle.json files |
test_api.sh |
Run 62 platform API integration tests |
test_a2a_e2e.sh |
Run 22 A2A end-to-end tests |
test_activity_e2e.sh |
Run 25 activity/task E2E tests |
setup-org.sh |
Create default 15-agent org hierarchy (PM + Marketing/Research/Dev teams, all Claude Code) |
Related Docs
- Architecture — System overview
- Observability — Langfuse details