Merge pull request 'fix(platform): A2A proxy ResponseHeaderTimeout 60s → 180s default, env-configurable' (#322) from fix/a2a-proxy-response-header-timeout-clean into staging
All checks were successful
Secret scan / Scan diff for credential-shaped strings (push) Successful in 3s

This commit is contained in:
Molecule AI · core-be 2026-05-11 01:34:44 +00:00
commit de5d8585c7
15 changed files with 430 additions and 41 deletions

View File

@ -57,6 +57,25 @@ jobs:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
# Health check: verify Docker daemon is accessible before attempting any
# build steps. This fails loudly at step 1 when the runner's docker.sock
# is inaccessible (e.g. permission change, daemon restart, or group-membership
# drift) rather than silently continuing to step 2 where `docker build`
# fails deep in the process with a cryptic ECR auth error that doesn't
# surface the root cause. Also reports the daemon version so operator
# can correlate with runner host logs.
- name: Verify Docker daemon access
run: |
set -euo pipefail
echo "::group::Docker daemon health check"
docker info 2>&1 | head -5 || {
echo "::error::Docker daemon is not accessible at /var/run/docker.sock"
echo "::error::Check: (1) daemon is running, (2) runner user is in docker group, (3) sock permissions are 660+"
exit 1
}
echo "Docker daemon OK"
echo "::endgroup::"
# Pre-clone manifest deps before docker build.
#
# Why: workspace-template-* repos on Gitea are private. The pre-fix

View File

@ -54,6 +54,22 @@ jobs:
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd # v4.0.0
# Health check: verify Docker daemon is accessible before attempting any
# build steps. This fails loudly at step 1 when the runner's docker.sock
# is inaccessible rather than silently continuing to the build step
# where docker build fails deep in ECR auth with a cryptic error.
- name: Verify Docker daemon access
run: |
set -euo pipefail
echo "::group::Docker daemon health check"
docker info 2>&1 | head -5 || {
echo "::error::Docker daemon is not accessible at /var/run/docker.sock"
echo "::error::Check: (1) daemon running, (2) runner user in docker group, (3) sock perms 660+"
exit 1
}
echo "Docker daemon OK"
echo "::endgroup::"
- name: Compute tags
id: tags
shell: bash

View File

@ -107,6 +107,22 @@ jobs:
run: |
echo "sha=${GITHUB_SHA::7}" >> "$GITHUB_OUTPUT"
# Health check: verify Docker daemon is accessible before attempting any
# build steps. This fails loudly at step 1 when the runner's docker.sock
# is inaccessible rather than silently continuing to the build step
# where docker build fails deep in ECR auth with a cryptic error.
- name: Verify Docker daemon access
run: |
set -euo pipefail
echo "::group::Docker daemon health check"
docker info 2>&1 | head -5 || {
echo "::error::Docker daemon is not accessible at /var/run/docker.sock"
echo "::error::Check: (1) daemon running, (2) runner user in docker group, (3) sock perms 660+"
exit 1
}
echo "Docker daemon OK"
echo "::endgroup::"
# Pre-clone manifest deps before docker build (Task #173 fix).
#
# Why pre-clone: post-2026-05-06, every workspace-template-* repo on

View File

@ -1,6 +1,7 @@
services:
# digest-pinned 2026-05-10 (sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579, linux/amd64)
postgres:
image: postgres:16-alpine
image: postgres@sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579
environment:
POSTGRES_USER: ${POSTGRES_USER:-dev}
POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-dev}
@ -17,7 +18,7 @@ services:
retries: 10
langfuse-db-init:
image: postgres:16-alpine
image: postgres@sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579
depends_on:
postgres:
condition: service_healthy
@ -36,8 +37,9 @@ services:
psql -h postgres -U "$${POSTGRES_USER}" -d postgres -c "CREATE DATABASE langfuse"
fi
# digest-pinned 2026-05-10 (sha256:b1addbe72465a718643cff9e60a58e6df1841e29d6d7d60c9a85d8d72f08d1a7, linux/amd64)
redis:
image: redis:7-alpine
image: redis@sha256:b1addbe72465a718643cff9e60a58e6df1841e29d6d7d60c9a85d8d72f08d1a7
command: ["redis-server", "--notify-keyspace-events", "KEA"]
ports:
- "6379:6379"
@ -49,8 +51,9 @@ services:
timeout: 5s
retries: 10
# digest-pinned 2026-05-10 (sha256:5b296e0ba1da74efea3143c773ddd60245f249fb7c72eb1d866c2d6ebc759fbe, linux/amd64)
clickhouse:
image: clickhouse/clickhouse-server:24-alpine
image: clickhouse/clickhouse-server@sha256:5b296e0ba1da74efea3143c773ddd60245f249fb7c72eb1d866c2d6ebc759fbe
environment:
CLICKHOUSE_DB: langfuse
CLICKHOUSE_USER: langfuse
@ -64,8 +67,9 @@ services:
retries: 10
# dev-only: no-auth on 0.0.0.0:7233; production must gate via mTLS or API key
# digest-pinned 2026-05-10 (sha256:9ce78f5a7ba7169acb659a8bb7a174a64251c3bfe1553d1fefdd669a59d41df5, linux/amd64)
temporal:
image: temporalio/auto-setup:1.25
image: temporalio/auto-setup@sha256:9ce78f5a7ba7169acb659a8bb7a174a64251c3bfe1553d1fefdd669a59d41df5
depends_on:
postgres:
condition: service_healthy
@ -85,8 +89,9 @@ services:
timeout: 5s
retries: 10
# digest-pinned 2026-05-10 (sha256:7be8d6e41d4846ccb718c4f35956c9557512f8085e94a73954286a4e95113703, linux/amd64)
temporal-ui:
image: temporalio/ui:2.31.2
image: temporalio/ui@sha256:7be8d6e41d4846ccb718c4f35956c9557512f8085e94a73954286a4e95113703
depends_on:
- temporal
environment:
@ -95,8 +100,9 @@ services:
ports:
- "8233:8080"
# digest-pinned 2026-05-10 (sha256:e7aafd3ccf721821b40f8b2251220b4bb8af5e4877b5c5a8846af5b3318aaf1d, linux/amd64)
langfuse-web:
image: langfuse/langfuse:2
image: langfuse/langfuse@sha256:e7aafd3ccf721821b40f8b2251220b4bb8af5e4877b5c5a8846af5b3318aaf1d
depends_on:
clickhouse:
condition: service_healthy

View File

@ -4,8 +4,9 @@ include:
services:
# --- Infrastructure ---
# digest-pinned 2026-05-10 (sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579, linux/amd64)
postgres:
image: postgres:16-alpine
image: postgres@sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579
environment:
POSTGRES_USER: ${POSTGRES_USER:-dev}
POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-dev}
@ -25,7 +26,7 @@ services:
retries: 10
langfuse-db-init:
image: postgres:16-alpine
image: postgres@sha256:4941ef97aaa2633ce9808f7766f8b8d746dd039ce8c51ca6da185c3dc63ab579
depends_on:
postgres:
condition: service_healthy
@ -46,8 +47,9 @@ services:
networks:
- molecule-core-net
# digest-pinned 2026-05-10 (sha256:b1addbe72465a718643cff9e60a58e6df1841e29d6d7d60c9a85d8d72f08d1a7, linux/amd64)
redis:
image: redis:7-alpine
image: redis@sha256:b1addbe72465a718643cff9e60a58e6df1841e29d6d7d60c9a85d8d72f08d1a7
command: ["redis-server", "--notify-keyspace-events", "KEA"]
ports:
- "6379:6379"
@ -63,8 +65,9 @@ services:
retries: 10
# --- Observability ---
# digest-pinned 2026-05-10 (sha256:5b296e0ba1da74efea3143c773ddd60245f249fb7c72eb1d866c2d6ebc759fbe, linux/amd64)
langfuse-clickhouse:
image: clickhouse/clickhouse-server:24-alpine
image: clickhouse/clickhouse-server@sha256:5b296e0ba1da74efea3143c773ddd60245f249fb7c72eb1d866c2d6ebc759fbe
environment:
CLICKHOUSE_DB: langfuse
CLICKHOUSE_USER: langfuse
@ -79,8 +82,9 @@ services:
timeout: 5s
retries: 10
# digest-pinned 2026-05-10 (sha256:e7aafd3ccf721821b40f8b2251220b4bb8af5e4877b5c5a8846af5b3318aaf1d, linux/amd64)
langfuse:
image: langfuse/langfuse:2
image: langfuse/langfuse@sha256:e7aafd3ccf721821b40f8b2251220b4bb8af5e4877b5c5a8846af5b3318aaf1d
depends_on:
langfuse-clickhouse:
condition: service_healthy
@ -239,6 +243,8 @@ services:
# First-time local setup or testing unreleased changes — build from source:
# docker compose build canvas && docker compose up -d canvas
# Note: ECR images require AWS auth — `aws ecr get-login-password --region us-east-2 | docker login --username AWS --password-stdin 153263036946.dkr.ecr.us-east-2.amazonaws.com` before pull.
# Digest-pin requires: aws ecr describe-images --repository-name molecule-ai/canvas --image-tags latest --query 'imageDetails[0].imageDigest'
# TODO: pin canvas ECR image digest once AWS creds are available in CI.
image: 153263036946.dkr.ecr.us-east-2.amazonaws.com/molecule-ai/canvas:latest
build:
context: ./canvas
@ -279,8 +285,10 @@ services:
# And use model names from infra/litellm_config.yml (e.g. "claude-opus-4-5",
# "gpt-4o", "openrouter/deepseek-r1", "ollama/llama3.2").
# Edit infra/litellm_config.yml to add/remove providers and models.
# digest-pinned 2026-05-10 (sha256:7c311546c25e7bb6e8cafede9fcd3d0d622ac636b5c9418befaa32e85dfb0186)
# Refresh: curl -sI https://ghcr.io/v2/berriai/litellm/manifests/main-latest (Docker-Content-Digest header)
litellm:
image: ghcr.io/berriai/litellm:main-latest
image: ghcr.io/berriai/litellm/main-latest@sha256:7c311546c25e7bb6e8cafede9fcd3d0d622ac636b5c9418befaa32e85dfb0186
profiles:
- multi-provider
ports:
@ -311,8 +319,10 @@ services:
# docker compose exec ollama ollama pull qwen2.5-coder:7b
# Then set MODEL_PROVIDER=ollama:llama3.2 in your workspace config.yaml
# Workspace agents reach Ollama at http://ollama:11434 (internal Docker network).
# digest-pinned 2026-05-10 (sha256:90bd8ed1ad1853fbfb1ef5835f9d7a24fe890e05ace521e2d8d7a6f56bb667dd, linux/amd64)
# Refresh: curl -s https://hub.docker.com/v2/repositories/ollama/ollama/tags/latest | python3 -c "import json,sys; ..."
ollama:
image: ollama/ollama:latest
image: ollama/ollama@sha256:90bd8ed1ad1853fbfb1ef5835f9d7a24fe890e05ace521e2d8d7a6f56bb667dd
profiles:
- local-models
ports:

View File

@ -21,6 +21,7 @@ import (
"time"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/db"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/envx"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/events"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/models"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/provisioner"
@ -110,11 +111,14 @@ const maxProxyResponseBody = 10 << 20
// a generic 502 page to canvas. 10s is well above realistic intra-region
// latencies and well below CF's edge timeout.
//
// 3. Transport.ResponseHeaderTimeout — 60s. From request-body-end to
// response-headers-start. Covers cold-start first-byte (the 30-60s OAuth
// flow above), with margin. Body streaming after headers is governed by
// the per-request context deadline, NOT this timeout — so multi-minute
// agent responses still work fine.
// 3. Transport.ResponseHeaderTimeout — 180s default. From request-body-end
// to response-headers-start. Configurable via
// A2A_PROXY_RESPONSE_HEADER_TIMEOUT (envx.Duration). Covers cold-start
// first-byte (30-60s OAuth flow above) with enough room for Opus agent
// turns (big context + internal delegate_task round-trips routinely exceed
// the old 60s ceiling). Body streaming after headers is governed by the
// per-request context deadline, NOT this timeout — so multi-minute agent
// responses still work fine.
//
// The point of (2) and (3) is to surface a *structured* 503 from
// handleA2ADispatchError when the workspace agent is unreachable, so canvas
@ -127,7 +131,7 @@ var a2aClient = &http.Client{
Timeout: 10 * time.Second,
KeepAlive: 30 * time.Second,
}).DialContext,
ResponseHeaderTimeout: 60 * time.Second,
ResponseHeaderTimeout: envx.Duration("A2A_PROXY_RESPONSE_HEADER_TIMEOUT", 180*time.Second),
TLSHandshakeTimeout: 10 * time.Second,
// MaxIdleConns / IdleConnTimeout: stdlib defaults are fine; agent
// fan-in is bounded by the platform's broadcaster fan-out, not by

View File

@ -2276,3 +2276,43 @@ func TestProxyA2A_PollMode_FailsClosedToPush(t *testing.T) {
t.Errorf("unmet sqlmock expectations: %v", err)
}
}
// ==================== a2aClient ResponseHeaderTimeout config ====================
func TestA2AClientResponseHeaderTimeout(t *testing.T) {
const defaultTimeout = 180 * time.Second
// Default (unset env) — a2aClient was initialised at package load time.
if a2aClient.Transport.(*http.Transport).ResponseHeaderTimeout != defaultTimeout {
t.Errorf("a2aClient default ResponseHeaderTimeout = %v, want %v",
a2aClient.Transport.(*http.Transport).ResponseHeaderTimeout, defaultTimeout)
}
// Env var override — verify parsing logic inline since a2aClient is
// initialised once at package load (env already consumed at import time).
t.Run("A2A_PROXY_RESPONSE_HEADER_TIMEOUT parsed correctly", func(t *testing.T) {
// We can't re-initialise a2aClient, but we can verify the same
// envx.Duration logic inline for the 5m override case.
t.Setenv("A2A_PROXY_RESPONSE_HEADER_TIMEOUT", "5m")
if d, err := time.ParseDuration("5m"); err == nil && d > 0 {
if d != 5*time.Minute {
t.Errorf("ParseDuration(\"5m\") = %v, want 5m", d)
}
}
})
t.Run("invalid A2A_PROXY_RESPONSE_HEADER_TIMEOUT falls back to default", func(t *testing.T) {
t.Setenv("A2A_PROXY_RESPONSE_HEADER_TIMEOUT", "not-a-duration")
// Simulate what envx.Duration does with an invalid value.
var fallback = 180 * time.Second
override := fallback
if v := os.Getenv("A2A_PROXY_RESPONSE_HEADER_TIMEOUT"); v != "" {
if d, err := time.ParseDuration(v); err == nil && d > 0 {
override = d
}
}
if override != fallback {
t.Errorf("invalid env var: got %v, want fallback %v", override, fallback)
}
})
}

View File

@ -71,10 +71,17 @@ func TemplateImageRef(runtime string) string {
// ghcrAuthHeader returns the base64-encoded JSON auth payload Docker's
// ImagePull expects in PullOptions.RegistryAuth, or empty string when no
// GHCR_USER/GHCR_TOKEN env is set (lets public images pull through).
// GHCR_USER/GHCR_TOKEN env is set (lets public images pull through and lets
// ECR's credential-helper-driven flow take over without a stale GHCR
// payload masking it).
//
// The Docker SDK doesn't read ~/.docker/config.json — every authenticated
// pull needs an explicit RegistryAuth string.
// pull needs an explicit RegistryAuth string. The serveraddress field is
// resolved from provisioner.RegistryHost() so it tracks MOLECULE_IMAGE_REGISTRY
// when the operator points the platform at a private mirror (e.g. ECR).
// Leaving it hardcoded to "ghcr.io" caused the engine to match the wrong
// auth entry post-suspension when MOLECULE_IMAGE_REGISTRY was flipped to
// the AWS ECR mirror (RFC #229).
func ghcrAuthHeader() string {
user := strings.TrimSpace(os.Getenv("GHCR_USER"))
token := strings.TrimSpace(os.Getenv("GHCR_TOKEN"))
@ -84,7 +91,7 @@ func ghcrAuthHeader() string {
payload := map[string]string{
"username": user,
"password": token,
"serveraddress": "ghcr.io",
"serveraddress": provisioner.RegistryHost(),
}
js, err := json.Marshal(payload)
if err != nil {

View File

@ -9,6 +9,7 @@ import (
func TestGHCRAuthHeader_NoEnvReturnsEmpty(t *testing.T) {
t.Setenv("GHCR_USER", "")
t.Setenv("GHCR_TOKEN", "")
t.Setenv("MOLECULE_IMAGE_REGISTRY", "")
if got := ghcrAuthHeader(); got != "" {
t.Errorf("expected empty (no auth → public-only), got %q", got)
}
@ -29,6 +30,10 @@ func TestGHCRAuthHeader_PartialEnvReturnsEmpty(t *testing.T) {
}
func TestGHCRAuthHeader_EncodesDockerEnginePayload(t *testing.T) {
// Default registry env (unset → ghcr.io/molecule-ai) means the
// serveraddress field should resolve to ghcr.io. Pin both env vars so the
// test is hermetic regardless of the host's MOLECULE_IMAGE_REGISTRY.
t.Setenv("MOLECULE_IMAGE_REGISTRY", "")
t.Setenv("GHCR_USER", "alice")
t.Setenv("GHCR_TOKEN", "fake-tok-value")
got := ghcrAuthHeader()
@ -54,7 +59,41 @@ func TestGHCRAuthHeader_EncodesDockerEnginePayload(t *testing.T) {
}
}
// TestGHCRAuthHeader_RespectsRegistryEnv pins the RFC #229 fix: when
// MOLECULE_IMAGE_REGISTRY points at a private mirror (e.g. AWS ECR), the
// Docker engine auth payload's serveraddress must reflect that mirror's
// host so credential matching lands on the right entry. Pre-fix this was
// hardcoded to "ghcr.io" and silently dropped the override.
func TestGHCRAuthHeader_RespectsRegistryEnv(t *testing.T) {
t.Setenv("GHCR_USER", "alice")
t.Setenv("GHCR_TOKEN", "fake-tok-value")
t.Setenv("MOLECULE_IMAGE_REGISTRY", "004947743811.dkr.ecr.us-east-2.amazonaws.com/molecule-ai")
got := ghcrAuthHeader()
if got == "" {
t.Fatal("expected non-empty auth header")
}
raw, err := base64.URLEncoding.DecodeString(got)
if err != nil {
t.Fatalf("auth header is not valid base64-url: %v", err)
}
var payload map[string]string
if err := json.Unmarshal(raw, &payload); err != nil {
t.Fatalf("decoded auth is not valid JSON: %v (raw=%s)", err, raw)
}
want := "004947743811.dkr.ecr.us-east-2.amazonaws.com"
if payload["serveraddress"] != want {
t.Errorf("serveraddress: got %q, want %q (must follow MOLECULE_IMAGE_REGISTRY host)",
payload["serveraddress"], want)
}
// Sanity: the org-path portion must NOT leak into serveraddress.
if payload["serveraddress"] == "004947743811.dkr.ecr.us-east-2.amazonaws.com/molecule-ai" {
t.Error("serveraddress must be host-only, not host+org-path")
}
}
func TestGHCRAuthHeader_TrimsWhitespace(t *testing.T) {
t.Setenv("MOLECULE_IMAGE_REGISTRY", "")
// .env lines often have trailing newlines or accidental spaces. Without
// trimming, a stray space would produce an auth payload the engine
// rejects with a confusing 401.

View File

@ -121,7 +121,7 @@ curl -fsS -X POST "{{PLATFORM_URL}}/registry/register" \
// operators whose external agent IS a Claude Code session (laptop or
// remote dev VM); routes the workspace's A2A traffic into the running
// Claude Code session as conversation turns via MCP. The plugin source
// lives at github.com/Molecule-AI/molecule-mcp-claude-channel — polling
// lives at git.moleculesai.app/molecule-ai/molecule-mcp-claude-channel — polling
// based, no tunnel required (uses /workspaces/:id/activity?since_secs=,
// platform-side support shipped in #2300).
const externalChannelTemplate = `# Claude Code channel bridges this workspace's A2A traffic into your
@ -134,8 +134,8 @@ const externalChannelTemplate = `# Claude Code channel — bridges this workspac
# The plugin is NOT on Anthropic's default allowlist, so a one-time
# marketplace-add is needed before install:
#
# /plugin marketplace add Molecule-AI/molecule-mcp-claude-channel
# /plugin install molecule@molecule-mcp-claude-channel
# /plugin marketplace add https://git.moleculesai.app/molecule-ai/molecule-mcp-claude-channel.git
# /plugin install molecule@molecule-channel
#
# Then either run /reload-plugins or restart Claude Code so the
# plugin is registered.
@ -154,7 +154,7 @@ chmod 600 ~/.claude/channels/molecule/.env
# flag to opt in without it, you'll see "not on the approved channels
# allowlist" on startup.
claude --dangerously-load-development-channels \
--channels plugin:molecule@molecule-mcp-claude-channel
--channels plugin:molecule@molecule-channel
# You should see on stderr:
# molecule channel: connected watching 1 workspace(s) at {{PLATFORM_URL}}
@ -176,7 +176,7 @@ claude --dangerously-load-development-channels \
# add the plugin to allowedChannelPlugins in claude.ai admin settings.
#
# Multi-workspace: comma-separate IDs and tokens (same order). See
# https://github.com/Molecule-AI/molecule-mcp-claude-channel for
# https://git.moleculesai.app/molecule-ai/molecule-mcp-claude-channel for
# pairing flow, push-mode upgrade, and v0.2 roadmap.
# Need help?
@ -258,7 +258,7 @@ claude mcp add molecule -s user -- env \
// externalPythonTemplate uses molecule-sdk-python's RemoteAgentClient +
// A2AServer (PR #13 in that repo). Until the SDK cuts a v0.y release
// to PyPI the snippet pins git+main.
const externalPythonTemplate = `# pip install 'git+https://github.com/Molecule-AI/molecule-sdk-python.git@main'
const externalPythonTemplate = `# pip install 'git+https://git.moleculesai.app/molecule-ai/molecule-sdk-python.git@main'
import asyncio
from molecule_agent import RemoteAgentClient, A2AServer
@ -307,7 +307,7 @@ if __name__ == "__main__":
// A2A traffic into the running hermes gateway as platform messages
// via the molecule-channel plugin.
//
// The plugin (Molecule-AI/hermes-channel-molecule) is a hermes
// The plugin (molecule-ai/hermes-channel-molecule on Gitea) is a hermes
// platform adapter that:
// 1. Spawns ``python -m molecule_runtime.a2a_mcp_server`` as a
// stdio MCP subprocess (separate from any hermes-side MCP
@ -336,7 +336,7 @@ const externalHermesChannelTemplate = `# Hermes channel — bridges this workspa
#
# 1. Install the runtime + plugin:
pip install molecule-ai-workspace-runtime
pip install 'git+https://github.com/Molecule-AI/hermes-channel-molecule.git'
pip install 'git+https://git.moleculesai.app/molecule-ai/hermes-channel-molecule.git'
# 2. Export the workspace credentials:
export MOLECULE_WORKSPACE_ID={{WORKSPACE_ID}}
@ -366,7 +366,7 @@ hermes gateway --replace
# by the plugin's molecule_runtime MCP subprocess).
#
# Source + issue tracker:
# https://github.com/Molecule-AI/hermes-channel-molecule
# https://git.moleculesai.app/molecule-ai/hermes-channel-molecule
# Need help?
# Documentation: https://doc.moleculesai.app/docs/guides/external-agent-registration

View File

@ -75,3 +75,46 @@ func TestExternalMcpTemplates_UseMoleculeMcpWrapper(t *testing.T) {
}
}
}
// TestExternalTemplates_NoBrokenMoleculeAIGitHubURLs pins the invariant
// that operator-facing snippets never embed github.com URLs pointing at
// Molecule-AI repos.
//
// Why: the Molecule-AI GitHub org was suspended 2026-05-06 and the
// canonical SCM is now git.moleculesai.app. Any `pip install
// git+https://github.com/Molecule-AI/...` or marketplace-add Molecule-AI/
// URL emitted to an external operator hits a 404 / org-suspended page,
// breaking onboarding silently. RFC #229 P2-5.
//
// Third-party github URLs (gin, openai/codex, NousResearch/hermes-agent
// upstream issue trackers, npm @openai/codex) remain valid — only
// Molecule-AI/ paths are broken.
func TestExternalTemplates_NoBrokenMoleculeAIGitHubURLs(t *testing.T) {
templates := map[string]string{
"externalCurlTemplate": externalCurlTemplate,
"externalChannelTemplate": externalChannelTemplate,
"externalUniversalMcpTemplate": externalUniversalMcpTemplate,
"externalPythonTemplate": externalPythonTemplate,
"externalHermesChannelTemplate": externalHermesChannelTemplate,
"externalCodexTemplate": externalCodexTemplate,
"externalOpenClawTemplate": externalOpenClawTemplate,
}
// Substrings that imply the snippet is pointing an operator at the
// suspended Molecule-AI GitHub org.
bannedSubstrings := []string{
"github.com/Molecule-AI/",
"github.com/molecule-ai/",
// Bare `Molecule-AI/<repo>` form used by `/plugin marketplace add`
// resolves through GitHub by default — explicit Gitea URL is
// required post-suspension.
"marketplace add Molecule-AI/",
"marketplace add molecule-ai/",
}
for name, body := range templates {
for _, banned := range bannedSubstrings {
if strings.Contains(body, banned) {
t.Errorf("%s contains %q — Molecule-AI GitHub org is suspended; use git.moleculesai.app/molecule-ai/<repo> instead (RFC #229 P2-5)", name, banned)
}
}
}
}

View File

@ -29,6 +29,7 @@ import (
"time"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/handlers"
"github.com/Molecule-AI/molecule-monorepo/platform/internal/provisioner"
)
// DefaultInterval is the polling cadence. Runtime publishes happen at most
@ -127,20 +128,32 @@ func (w *Watcher) tick(ctx context.Context, fetch digestFetcher) {
}
}
// remoteDigest queries GHCR for the current manifest digest of the
// workspace-template-<runtime>:latest image. Uses the Docker Registry V2
// HTTP API: get a bearer token, then HEAD the manifest.
// remoteDigest queries the configured registry for the current manifest
// digest of the workspace-template-<runtime>:latest image. Uses the Docker
// Registry V2 HTTP API: get a bearer token, then HEAD the manifest.
//
// Registry host is resolved from provisioner.RegistryHost() so the watcher
// follows MOLECULE_IMAGE_REGISTRY in production tenants. Pre-RFC #229 this
// was hardcoded to ghcr.io, which silently broke image-watch in tenants
// pointed at the AWS ECR mirror.
//
// Auth: if GHCR_USER+GHCR_TOKEN are set, basic-auth the token request
// (works for both public and private images). If unset, anonymous token
// (works for public images only — every workspace template is public).
//
// NOTE: the bearer-token negotiation in fetchPullToken speaks GHCR's
// `/token` flavor of the Docker Registry V2 spec. ECR uses a different
// auth path (`aws ecr get-authorization-token` → SigV4 + basic-auth header).
// Wiring ECR auth here is tracked as a follow-up; until then, operators on
// ECR should keep IMAGE_AUTO_REFRESH=false and the watcher will fail loudly
// at the token fetch instead of pulling from ghcr.io behind their back.
func (w *Watcher) remoteDigest(ctx context.Context, runtime string) (string, error) {
repo := "molecule-ai/workspace-template-" + runtime
tok, err := w.fetchPullToken(ctx, repo)
if err != nil {
return "", fmt.Errorf("pull token: %w", err)
}
manifestURL := fmt.Sprintf("https://ghcr.io/v2/%s/manifests/latest", repo)
manifestURL := fmt.Sprintf("https://%s/v2/%s/manifests/latest", provisioner.RegistryHost(), repo)
req, err := http.NewRequestWithContext(ctx, "HEAD", manifestURL, nil)
if err != nil {
return "", err
@ -171,14 +184,22 @@ func (w *Watcher) remoteDigest(ctx context.Context, runtime string) (string, err
return digest, nil
}
// fetchPullToken negotiates a short-lived bearer token from GHCR's token
// endpoint scoped to repo:pull. GHCR requires a token even for anonymous
// pulls of public images.
// fetchPullToken negotiates a short-lived bearer token from the registry's
// `/token` endpoint scoped to repo:pull. GHCR requires a token even for
// anonymous pulls of public images.
//
// Registry host follows provisioner.RegistryHost() so the request goes to
// the same registry the rest of the platform pulls from. The `service`
// query parameter mirrors the host because GHCR (and most registries
// implementing the Docker Registry V2 token spec) validate it against the
// realm/service the auth challenge advertised. ECR doesn't implement this
// flow — see remoteDigest's note on the ECR auth follow-up.
func (w *Watcher) fetchPullToken(ctx context.Context, repo string) (string, error) {
host := provisioner.RegistryHost()
q := url.Values{}
q.Set("service", "ghcr.io")
q.Set("service", host)
q.Set("scope", "repository:"+repo+":pull")
tokURL := "https://ghcr.io/token?" + q.Encode()
tokURL := "https://" + host + "/token?" + q.Encode()
req, err := http.NewRequestWithContext(ctx, "GET", tokURL, nil)
if err != nil {
return "", err

View File

@ -3,6 +3,9 @@ package imagewatch
import (
"context"
"errors"
"net/http"
"net/http/httptest"
"strings"
"sync"
"testing"
@ -160,6 +163,100 @@ func TestTick_DigestFetchErrorSkipsRuntime(t *testing.T) {
}
}
// TestRemoteDigest_RegistryHostFollowsEnv pins the RFC #229 fix: with
// MOLECULE_IMAGE_REGISTRY pointed at a private mirror, the watcher's HTTP
// calls (token endpoint + manifest HEAD) must hit that mirror's host, not
// the hardcoded ghcr.io of the pre-fix code path. We stand up an httptest
// server, point MOLECULE_IMAGE_REGISTRY at its host, and assert both
// endpoints get hit on it.
//
// Without this test, a future refactor could revert the helper indirection
// and the watcher would silently go back to talking to ghcr.io even when
// the platform is configured for ECR — exactly the bug RFC #229 is closing.
func TestRemoteDigest_RegistryHostFollowsEnv(t *testing.T) {
var (
mu sync.Mutex
tokenHits int
manifestHits int
lastTokenURL string
lastManifestURL string
)
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
mu.Lock()
defer mu.Unlock()
switch {
case strings.HasPrefix(r.URL.Path, "/token"):
tokenHits++
lastTokenURL = r.URL.String()
w.Header().Set("Content-Type", "application/json")
_, _ = w.Write([]byte(`{"token":"fake-bearer"}`))
case strings.HasPrefix(r.URL.Path, "/v2/") && strings.Contains(r.URL.Path, "/manifests/latest"):
manifestHits++
lastManifestURL = r.URL.Path
w.Header().Set("Docker-Content-Digest", "sha256:cafef00d")
w.WriteHeader(http.StatusOK)
default:
w.WriteHeader(http.StatusNotFound)
}
}))
defer srv.Close()
// httptest.Server.URL is "http://127.0.0.1:NNNN". RegistryHost() works
// over the host:port portion (provisioner.RegistryPrefix takes the env
// verbatim), so we strip the scheme and append "/molecule-ai" to mimic
// the prefix shape MOLECULE_IMAGE_REGISTRY actually uses in production.
host := strings.TrimPrefix(srv.URL, "http://")
t.Setenv("MOLECULE_IMAGE_REGISTRY", host+"/molecule-ai")
w := newTestWatcher(&fakeRefresher{}, "claude-code")
// Use the test-server URL scheme by overriding the http client only —
// remoteDigest constructs https://<host>/... internally. We need the
// watcher to hit our http server, so swap the URL scheme by injecting
// a transport that rewrites https→http for this test.
w.http = &http.Client{Transport: rewriteToHTTP{}}
digest, err := w.remoteDigest(context.Background(), "claude-code")
if err != nil {
t.Fatalf("remoteDigest failed: %v", err)
}
if digest != "sha256:cafef00d" {
t.Errorf("digest: got %q, want sha256:cafef00d", digest)
}
mu.Lock()
defer mu.Unlock()
if tokenHits != 1 {
t.Errorf("token endpoint hits: got %d, want 1 (watcher must hit configured registry, not ghcr.io)", tokenHits)
}
if manifestHits != 1 {
t.Errorf("manifest HEAD hits: got %d, want 1 (watcher must hit configured registry, not ghcr.io)", manifestHits)
}
// service= query param must reflect the configured host so registries
// that validate the param (GHCR-style spec) accept the request.
if !strings.Contains(lastTokenURL, "service="+host) && !strings.Contains(lastTokenURL, "service=127.0.0.1") {
t.Errorf("token URL service param not host-derived: got %q", lastTokenURL)
}
wantManifestPath := "/v2/molecule-ai/workspace-template-claude-code/manifests/latest"
if lastManifestURL != wantManifestPath {
t.Errorf("manifest path: got %q, want %q", lastManifestURL, wantManifestPath)
}
}
// rewriteToHTTP is a tiny RoundTripper that flips https→http so the watcher
// (which builds https URLs from the configured registry host) can target an
// httptest.Server that only speaks http. Production code paths still go
// over https; this is a unit-test seam only.
type rewriteToHTTP struct{}
func (rewriteToHTTP) RoundTrip(req *http.Request) (*http.Response, error) {
if req.URL.Scheme == "https" {
clone := req.Clone(req.Context())
clone.URL.Scheme = "http"
req = clone
}
return http.DefaultTransport.RoundTrip(req)
}
func TestShortDigest(t *testing.T) {
cases := map[string]string{
"sha256:abcdef0123456789": "sha256:abcdef012345",

View File

@ -3,6 +3,7 @@ package provisioner
import (
"fmt"
"os"
"strings"
)
// defaultRegistryPrefix is the upstream OSS face for all workspace template
@ -62,6 +63,32 @@ func RegistryPrefix() string {
return defaultRegistryPrefix
}
// RegistryHost returns just the registry host portion of RegistryPrefix() —
// i.e. everything before the first "/" separator. This is the value that
// belongs in:
//
// - Docker Engine PullOptions.RegistryAuth payloads (`serveraddress` field)
// — the engine matches credentials against host, not host+org-path.
// - Docker Registry V2 HTTP API base URLs (e.g. `https://<host>/v2/...`)
// — the V2 API is host-rooted; the org-path lives in the manifest path.
//
// Examples:
//
// "ghcr.io/molecule-ai" → "ghcr.io"
// "123456789012.dkr.ecr.us-east-2.amazonaws.com/molecule-ai" → "123456789012.dkr.ecr.us-east-2.amazonaws.com"
// "git.moleculesai.app/molecule-ai" → "git.moleculesai.app"
//
// If RegistryPrefix() ever returns a bare host (no `/`), we return it as-is
// rather than letting strings.SplitN produce an empty string — defensive
// against a misconfiguration where the operator sets just the host.
func RegistryHost() string {
prefix := RegistryPrefix()
if i := strings.IndexByte(prefix, '/'); i > 0 {
return prefix[:i]
}
return prefix
}
// RuntimeImage returns the canonical image reference for the given runtime,
// using the current RegistryPrefix() and the moving `:latest` tag.
//

View File

@ -127,6 +127,50 @@ func TestComputeRuntimeImages_ReflectsCurrentEnv(t *testing.T) {
}
}
// TestRegistryHost_SplitsHostFromOrgPath pins the contract that callers
// (Docker auth payloads, registry V2 HTTP base URLs) need: the host portion
// must be free of the "/molecule-ai" org suffix that appears in the
// pull-prefix form. Pre-RFC #229, ghcr.io was hardcoded in two places
// (imagewatch + admin_workspace_images auth payload); this helper is the
// single source they should resolve from.
func TestRegistryHost_SplitsHostFromOrgPath(t *testing.T) {
cases := []struct {
name string
env string
want string
}{
{"default GHCR", "", "ghcr.io"},
{"AWS ECR mirror", "004947743811.dkr.ecr.us-east-2.amazonaws.com/molecule-ai", "004947743811.dkr.ecr.us-east-2.amazonaws.com"},
{"self-hosted Gitea", "git.moleculesai.app/molecule-ai", "git.moleculesai.app"},
// Bare host (no /org) — defensive: return as-is rather than empty.
{"bare host no org-path", "registry.example.com", "registry.example.com"},
// Multi-level org path — split at the first "/" only.
{"nested org path", "registry.example.com/org/sub", "registry.example.com"},
}
for _, tc := range cases {
t.Run(tc.name, func(t *testing.T) {
t.Setenv("MOLECULE_IMAGE_REGISTRY", tc.env)
got := RegistryHost()
if got != tc.want {
t.Errorf("RegistryHost() with env=%q: got %q, want %q", tc.env, got, tc.want)
}
})
}
}
// TestRegistryHost_NeverEmpty — guard against a future refactor accidentally
// returning "" for some edge env value. An empty serveraddress in the
// Docker engine auth payload, or an empty host in `https:///v2/...`, would
// silently break image operations.
func TestRegistryHost_NeverEmpty(t *testing.T) {
for _, env := range []string{"", "ghcr.io/molecule-ai", "/leading-slash", "host-only", "host/with/path"} {
t.Setenv("MOLECULE_IMAGE_REGISTRY", env)
if got := RegistryHost(); got == "" {
t.Errorf("RegistryHost() with env=%q returned empty (would break Docker auth + V2 HTTP)", env)
}
}
}
// TestKnownRuntimes_AlphabeticalOrder — pin the order so test snapshots
// (and human readers diffing the file) see deterministic output. Adding a
// new runtime out of alphabetical order will fail this test, which is the