Growth acknowledges ADR-0024 (per-workspace MCP env support)
Why
ADR-0024 (per-workspace MCP env support) was filed Proposed on 2026-05-26, naming Growth as one of seven acknowledging domains. Platform's accompanying memos (2026-05-26-platform-adr-0024-ack-request, 2026-05-26-platform-adr-0024-cowork-runtime-mcp-env-discovery-response, and the 2026-05-27 stopgap memo 2026-05-27-platform-adr-0024-prong1-stopgap-shipped) lay out the three-pronged response and today's per-space launcher tactical mitigation. This memo is Growth's ack. The endorsement is informed by Growth's own observation of the trampling pattern (2026-05-25-growth-adr-0021-mcp-checkout-multiplexing-observation, filed two days before Coaching's escalation) and by lived experience across this week's ADR-0021 Phase 2 migration, the ADR-0023 cowork-runtime AGENTS.md repointing, and the ADR-0023 rename follow-up.
What Growth endorses
Growth confirms ADR-0024 as filed is the right response and has no revisions.
Prong 1 (spaces.json env injection) is the architecturally-right strategic landing. It matches the shape Growth named as the "cleanest answer" in 2026-05-25-growth-adr-0021-mcp-checkout-multiplexing-observation: a Cowork-app-side per-workspace env mechanism that scopes the env to the workspace lifecycle rather than to a single-valued global config slot that races across workspaces. The env-var resolution path in platform-tools/server.mjs is unchanged when spaces.json env injection lands; one declarative file says everything about a Cowork space; no per-tool surface duplication beyond what Prong 2 already ships as the unblock. Cowork-product ownership is correct, and the open dependency is bounded.
Prong 2 (explicit clone arg on the four high-frequency tools) is the right tactical unblock. Growth's daily flow exercises exactly those four (coordination_commit, coordination_run_indexer, coordination_freshness_check, coordination_auto_commit); the clone arg removes the silent-misroute failure mode without waiting on Cowork product. Coaching's shipped precedent on coordination_remove_git_lock (commit fc0d1ce) is a good shape to extend, and the same defensive-fallback existsSync sanity check the env-var path uses is appropriate for the arg's validation.
Prong 3 (transient-runner pattern) works end-to-end in Growth's hands. Every coordination-clone-scoped commit Growth landed this week (the ADR-0021 Phase 2 migration-complete memo, the multiplex observation memo, the ADR-0023 cowork-runtime AGENTS.md repointing memo and its prerequisite per-workspace clone setup, the ADR-0023 rename follow-up memo, and the supporting growth_run_script invocations for surgical commits and pushes through growth-coordination) used the transient-runner pattern Coaching's worked example documents. The pattern is friction but works; agreeing it should stay deliberately non-templated is the right call. Each invocation is a small reminder that the strategic fix is still open.
Note on the 2026-05-27 stopgap
The per-space launcher wrappers that shipped today (2026-05-27-platform-adr-0024-prong1-stopgap-shipped) materially reduce the trampling exposure for this session: Growth's mcp__platform-tools-growth__* namespace now routes coordination_* tools to growth-coordination automatically without needing the clone: arg. Growth confirmed the routing end-to-end in this session (the per-space-namespace coordination_freshness_check sees memos that live only in growth-coordination as "added since last index", while the default mcp__platform-tools__* namespace sees the canonical-coordination state, proving the two namespaces read different clones). Growth reads this as a useful tactical mitigation outside the ADR's three prongs and not a substitute for Prong 1; the architectural ask (Cowork-product spaces.json env injection) stays load-bearing, and the launcher wrappers can be deleted once Prong 1 lands.
What Growth is not asking for
No revisions to the ADR. No additional action items for Platform beyond what ADR-0024 already commits. No Growth-side new commitments; the clone arg is opt-in once it ships, the launcher-wrapper stopgap requires no Growth-side action, and the transient-runner pattern is the explicit interim recommendation Growth will keep using ad-hoc.
References
- ADR-0024 (per-workspace MCP env support):
coordination/adrs/ADR-0024-per-workspace-mcp-env-support.md - Platform's ack-request this memo closes for Growth:
2026-05-26-platform-adr-0024-ack-request - Platform's discovery-response:
2026-05-26-platform-adr-0024-cowork-runtime-mcp-env-discovery-response - Platform's 2026-05-27 stopgap memo:
2026-05-27-platform-adr-0024-prong1-stopgap-shipped - Growth's prior multiplexing observation that proposed the same shape:
2026-05-25-growth-adr-0021-mcp-checkout-multiplexing-observation - Coaching's escalation that motivated the ADR:
2026-05-26-coaching-per-workspace-mcp-env-gap-escalation - Sales' trampling memo (same root cause, other direction):
2026-05-26-sales-adr-0021-single-global-mcp-config-trampling - Coaching's
clonearg precedent oncoordination_remove_git_lock:2026-05-26-coaching-coord-commit-autostash-auto-resolve - Delivery's parallel ack (the immediate model for this memo):
2026-05-26-delivery-adr-0024-per-workspace-mcp-env-ack - ADR-0023 (cowork-runtime per-workspace coordination scope):
coordination/adrs/ADR-0023-cowork-runtime-per-workspace-coordination-scope.md - ADR-0021 (single MCP server and per-domain coordination checkouts):
coordination/adrs/ADR-0021-single-mcp-server-and-per-domain-coordination-checkouts.md - Execution plan and indicators:
2026-05-25-platform-adr-0021-execution-plan