@terreno/mcp 57.4.0 → 57.6.0

package/dist/docs/versioned/next/scans/reduce-cold-compile-time/charter.md

@@ -0,0 +1,129 @@
+# Scan charter — reduce cold `bun run compile` wall-clock time
+
+- Slug: reduce-cold-compile-time
+- Status: approved (2026-09-20)
+- Owner: joshgachnang
+
+## Goal
+
+The terreno monorepo compiles with per-package standalone `tsc` and no build
+cache, TS project references, or shared-dep deduplication. A full cold
+`bun run compile` currently takes ~305s, and the pipeline recompiles shared
+packages (`@terreno/test`, `@terreno/api`, `@terreno/syncdb`, `@terreno/jobs`)
+multiple times per run. This campaign cuts cold compile wall-clock by 30% by
+removing redundant tsc passes, deduplicating shared-dependency recompiles,
+introducing project references, and trimming non-typecheck work from the compile
+path — without changing published dist output or weakening type-safety. Faster
+compiles shorten every bootstrap, CI run, and agent iteration.
+
+## Metric
+
+| Field | Value |
+| --- | --- |
+| Name | Cold full-monorepo compile wall-clock |
+| Command | `find . -maxdepth 2 -name '*.tsbuildinfo' -not -path '*/node_modules/*' -delete; rm -rf */dist; time bun run compile` (min of 3 runs, seconds) |
+| Unit | s |
+| Direction | down |
+| Protocol | Cold (dist + tsbuildinfo removed before each run), best (min) of 3 consecutive runs, same VM |
+| Baseline | 305 (measured 2026-09-20) |
+| Target | 214 |
+| Horizon | 4 rounds |
+
+Baseline evidence:
+
+```text
+run1 exit=0 elapsed=321s
+run2 exit=0 elapsed=305s
+run3 exit=0 elapsed=309s
+# best of 3 = 305s ; target -30% = 214s
+```
+
+## Scope
+
+- In scope: build/compile configuration only — root `package.json` compile
+  scripts, per-package `tsconfig.json` / `tsconfig.server.json`,
+  `.github/scripts/compile-workspace-deps.js`, per-package `compile` scripts,
+  and any new build tooling that measurably reduces cold compile time.
+- Out of scope: product/source logic, test assertions, runtime behavior,
+  dependency version bumps unrelated to the build.
+- Unacceptable trade-offs:
+  - **Published dist output must stay intact** — `declaration` (.d.ts) emit and
+    the published `dist/` layout for every `@terreno/*` package are preserved.
+  - **No type-safety loss** — no loosening `strict`/`strictNullChecks`, no
+    removing existing `skipLibCheck`, no `noEmit`-to-skip-checks tricks. Adding
+    `skipLibCheck` where absent is allowed.
+  - Publish-time behavior of `compile-workspace-deps.js` must be preserved
+    (it exists because tag-publish replaces `workspace:*` refs before install).
+
+New build dependencies ARE permitted (owner did not exclude them), provided the
+two guardrails above hold.
+
+## Detection rules
+
+| Rule | Query | Matches | Why it moves the metric | Hits |
+| --- | --- | --- | --- | --- |
+| redundant-pipeline | Inspect root `compile` script; `grep -rl compile-workspace-deps */package.json` | Root `compile` runs `@terreno/test` and `@terreno/api` explicitly, then `--filter '*'` recompiles them again; `api` and `admin-backend` re-`tsc` shared deps via `compile-workspace-deps.js` | Every duplicate tsc pass of a large package (api=90k LOC, syncdb, test) is pure cold-time waste | 4 |
+| no-project-references | For each compile-path tsconfig: `grep -L '"references"'`; check for a root solution tsconfig | 19 tsconfigs compile as isolated `tsc` with no `references`; no root solution tsconfig | Project references (`tsc -b`) compile each package once in dep order and enable parallelism, killing recompile overlap | 19 |
+| non-typecheck-work-in-compile | `grep -E 'sync-\|cp \|&&.*&&'` in each package `compile` script | `mcp-server` runs 3 doc-sync scripts at 8GB heap plus a `cp` before/around its `tsc` on every compile | Non-tsc work on the compile critical path adds fixed cold-time cost unrelated to type-checking | 1 |
+| missing-skiplibcheck | For each compile-participating tsconfig: `grep '"skipLibCheck"'` | `demo/tsconfig.json` and `website/tsconfig.json` lack `skipLibCheck` (their compile scripts use these configs) | Without `skipLibCheck`, tsc type-checks every `.d.ts` in the dependency tree — costly for RN/Expo/docs dep trees | 2 |
+
+## Exclusions
+
+Beyond the map-reduce defaults: `node_modules/`, `*/dist/`, `patches/`,
+`terraform/`, generated files (`openApiSdk.ts`, `plugins/terreno-claude/`), and
+`example-frontend` (no `compile` script — not on the compile path).
+
+## Validity rule
+
+A finding is valid only when applying it measurably reduces cold `bun run compile`
+wall-clock (best-of-3) while keeping published dist output identical and
+type-safety unchanged.
+
+## Severity rubric
+
+| Severity | Threshold |
+| --- | --- |
+| high | Plausibly removes ≥30s of cold compile time |
+| medium | Plausibly removes 5–30s |
+| low | Plausibly removes <5s |
+
+Effort: S = under an hour, M = under a day, L = more than a day or needs design.
+
+## Slice policy
+
+- Maximum 4 files and 3 findings per slice.
+- Never mix: a project-references migration with unrelated tsconfig flag tweaks;
+  or `compile-workspace-deps.js` changes with anything else.
+- Always its own slice: the `tsc -b` / project-references migration (touches
+  every package + CI; reviewer-heavy) and any `compile-workspace-deps.js`
+  change (publish-time behavior risk).
+
+## Review routing
+
+| Field | Value |
+| --- | --- |
+| Mode | fixed |
+| Reviewers | @joshgachnang |
+| Assignee | @joshgachnang |
+| Teams | — |
+| Draft | no |
+| Labels | reduce-cold-compile-time, build-perf |
+| Merge policy | human |
+| Notify | PR body mention |
+
+## Budget
+
+- Rounds: 4
+- Slices per round: 2
+- Open PRs at once (`wipLimit`): 1
+- Minimum metric step per round before reporting diminishing returns: 10s
+
+## Decisions
+
+| Decision | Question that prompted it |
+| --- | --- |
+| Goal = reduce cold `bun run compile` wall-clock | "What goal should this scan campaign drive toward?" |
+| Target = 214s (−30% of 305s baseline) | "How aggressive a cold-compile reduction should this campaign target?" |
+| Guardrails = keep published dist output + no type-safety loss; new build deps allowed | "What must this campaign NOT break while cutting compile time?" |
+| Review = fixed @joshgachnang, human merge | "Who reviews the campaign's PRs, and what's the merge policy?" |
+| Budget = 1 open PR, 2 slices/round, 4 rounds | "How much work-in-progress should the campaign carry per round?" |

package/dist/docs/versioned/next/scans/reduce-cold-compile-time/round-1-findings.md

@@ -0,0 +1,62 @@
+# Round 1 findings — reduce cold `bun run compile` time
+
+- Goal: reduce cold `bun run compile` wall-clock across the terreno monorepo
+- Metric: cold compile seconds (best of 3). **Baseline 305s → target 214s (−30%)**
+- Head: `f8779a96`
+- Kept: 4 findings · Merged: 2 candidates · Dropped: 2 candidates
+
+## Measured evidence (Sift verification)
+
+| Measurement | Result |
+| --- | --- |
+| `@terreno/test` compile (cold) | 10s |
+| `@terreno/api` compile FULL (compile-workspace-deps + tsc, cold) | 50s |
+| `@terreno/api` tsc-only (deps prebuilt) | 31s → compile-workspace-deps adds **~19s per api compile** |
+| demo `tsc --noEmit` without vs with `--skipLibCheck` | 10s vs 10s → **0s** delta |
+| mcp-server sync scripts (sync-ui / sync-versioned / sync-package) | 0s + 5s + 0s = **~5s** |
+
+## Ranked findings
+
+| # | ID | Rule | Sev | Conf | Effort | Est. cold-time saving | Anchors |
+| --- | --- | --- | --- | --- | --- | --- | --- |
+| 1 | redundant-explicit-prefix | redundant-pipeline | high | high | S* | ~60s (1× test + 1× api) | `package.json:223` |
+| 2 | compile-workspace-deps-overlap | redundant-pipeline | high | high | M | ~19s+ per api compile, ×2–3 dependents | `.github/scripts/compile-workspace-deps.js:22`, `api/package.json:117`, `admin-backend/package.json:47` |
+| 3 | no-project-references | no-project-references | high | medium | L | large but unquantified; durable fix that enables #1/#2 safely | `api/tsconfig.json:1` + 18 more; no root solution tsconfig |
+| 4 | mcp-doc-sync-on-compile-path | non-typecheck-work-in-compile | medium | high | S | ~5s | `mcp-server/package.json:52` |
+
+\* #1 is a one-line edit but is **load-bearing for build ordering** — safe only once #2/#3 provide dependency-ordered builds. See coupling note below.
+
+## High-severity findings (adversarial verification)
+
+### 1. Redundant explicit `test`+`api` prefixes in root compile — CONFIRMED
+Root `compile` = `bun run --filter '@terreno/test' compile && bun run --filter '@terreno/api' compile && bun run --filter '*' compile`. The final `--filter '*'` already builds all 19 workspaces including `test` and `api`, so the two prefixes are a full extra cold compile of test (10s) + api (50s) = **~60s of duplication**.
+
+*Refutation attempted:* "The prefixes force build order; removing them breaks `*`." → Partially holds. `compile-workspace-deps.js` makes api self-build its deps, so api compiles fine inside `*`; but bun `--filter '*'` gives no guaranteed dependency order, so the prefixes currently serve as a crude ordering hack. Verdict: the **duplication is real and confirmed**, but the safe fix is to introduce real ordering (#2/#3), not to blindly delete the prefixes. Kept high.
+
+### 2. compile-workspace-deps.js re-tsc's shared deps every compile — CONFIRMED
+`compile-workspace-deps.js:22` (`const compiled = new Set()`) dedupes only within one node process. Each package's compile runs in its own `bun run` process, so api re-builds test+syncdb (+19s measured over tsc-only), and admin-backend re-builds api (~50s!)+jobs+test — while `--filter '*'` also builds those same packages as their own targets. Shared deps are cold-compiled multiple times per run.
+
+*Refutation attempted:* "The script is required so deps' `dist` exists before dependents compile." → Holds for the current unordered `*` build, and the script's own doc comment says it exists for tag-publish (post-`workspace:*`-replacement installs). Verdict: confirmed redundant **for the dev/monorepo build**; the fix must preserve publish-time behavior (charter guardrail) and supply ordering another way. Kept high, effort M.
+
+### 3. No TS project references / `tsc -b` solution build — CONFIRMED
+No package tsconfig has `references` or `composite`, and there is no root solution tsconfig (`grep -c '"composite"'` → 0 everywhere; `ls tsconfig.json` → not found). Every package is a full standalone `tsc`.
+
+*Refutation attempted:* "Project references help warm builds (via .tsbuildinfo reuse), not cold — this campaign measures cold." → Fails: even cold, `tsc -b` walks the reference graph and compiles each package **exactly once** in dependency order and parallelizes independent leaves, eliminating the multi-recompile waste that #1 and #2 describe. It is the durable structural fix that makes #1 and #2 safe. Verdict: confirmed high; magnitude unproven without doing the migration → confidence medium, effort L.
+
+## Coupling note (for Plot)
+
+Findings #1, #2, #3 are one problem at three depths: the pipeline recompiles shared packages because there is no dependency-ordered build. #3 (project references) is the durable fix and would subsume #1/#2. A lighter interim fix is bun topological filter ordering. Per charter slice policy: the project-references migration and any `compile-workspace-deps.js` change each get **their own slice**. Finding #4 is fully independent and is the cheapest standalone win.
+
+## Merged candidates
+
+- `s1-pipeline-3` ("test compiled 3×") — folded into #1 and #2; it is the same duplication counted from test's side.
+- `s3-mcpwork-2` ("8GB heap flag on copy scripts") — folded into #4; negligible sub-part of the same compile string.
+
+## Dropped candidates
+
+| Candidate | Reason |
+| --- | --- |
+| `s4-skiplibcheck-1` (demo) | **Invalid — refuted.** `skipLibCheck` is already inherited from `expo/tsconfig.base` (`skipLibCheck: true`). Measured delta 10s vs 10s = 0s. No metric movement. |
+| `s4-skiplibcheck-2` (website) | **Invalid — refuted.** `skipLibCheck` already inherited from `@docusaurus/tsconfig` (`skipLibCheck: true`). `--noEmit` + tiny source set → no measurable movement. |
+
+Drop reasons: 2/2 = already-satisfied-via-inheritance / no-metric-movement.

package/dist/docs/versioned/next/tasks/agent-ui-blocks.md

@@ -5,7 +5,7 @@
 
 Every task is a vertical slice: contract + renderer or producer + docs + Bun tests. Work
 the frontier (blockers complete). Phase 2 chart/table work is additionally blocked on
-[PR #1302](https://github.com/FlourishHealth/terreno/pull/1302) merging to `master`.
+[PR #1302](https://github.com/TerrenoLabs/terreno/pull/1302) merging to `master`.
 
 ### Phase 1: Contract — `@terreno/blocks`
 

package/dist/docs/versioned/next/tasks/better-auth-strict-oauth-provider.md

@@ -2,7 +2,7 @@
 
 **Feature profile:** true  
 **IP:** [better-auth-strict-oauth-provider.md](../implementationPlans/better-auth-strict-oauth-provider.md)  
-**Issue:** https://github.com/FlourishHealth/terreno/issues/1218
+**Issue:** https://github.com/TerrenoLabs/terreno/issues/1218
 
 ## Durable contract
 

package/dist/docs/versioned/next/tasks/charts-and-dashboards.md

@@ -1,7 +1,7 @@
 # Tasks: Charts and dashboard primitives
 
 IP: [charts-and-dashboards.md](../implementationPlans/charts-and-dashboards.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1175
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1175
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/compile-pipeline-dedup.md

@@ -0,0 +1,61 @@
+# Tasks: Compile pipeline dedup
+
+IP: [compile-pipeline-dedup.md](../implementationPlans/compile-pipeline-dedup.md)
+Slice: pipeline-ordering-dedup · Scan goal: reduce-cold-compile-time (305s → 214s)
+
+Tracer-bullet order. Each task names files, acceptance, and verification.
+
+## T0 — Capture baseline artifacts (blocking, no code change)
+
+- **Files:** none (read-only).
+- **Do:** `rm -rf */dist && bun run compile 2>&1 | tee /tmp/baseline-build.log`; copy every
+  package's `dist/` into a baseline snapshot dir for later `diff -r`.
+- **Acceptance:** baseline build log + dist snapshot captured; note the cold time (best of
+  3 already recorded = 305s).
+- **Verify:** snapshot dir exists with per-package `dist/`.
+
+## T1 — Apply the three compile-script edits
+
+- **Files:** `package.json`, `api/package.json`, `admin-backend/package.json`.
+- **Do:**
+  - root `compile` → `bun run --filter '*' compile`
+  - `api` `compile` → `bun tsc`
+  - `admin-backend` `compile` → `bun tsc`
+- **Acceptance:** the three scripts match the IP; `.github/scripts/compile-workspace-deps.js`
+  is untouched (`git diff --stat` excludes it).
+- **Blockers:** none. **Docs:** none yet.
+
+## T2 — Prove correctness: single-build, clean, dist-identical
+
+- **Files:** none (verification).
+- **Do:** `rm -rf */dist && bun run compile 2>&1 | tee /tmp/new-build.log`; then
+  `diff -r` each package's new `dist/` against the T0 snapshot.
+- **Acceptance (IP criteria 1–3):**
+  - exit 0 from clean tree
+  - each buildable `@terreno/*` package appears once; **zero** `Compiling @terreno/... with
+    bun tsc` lines (that string only comes from `compile-workspace-deps.js`, which must not
+    run during the dev build)
+  - `diff -r` reports no differences in any `dist/`
+- **Verify:** grep `/tmp/new-build.log` for per-package compile count and for
+  `compile-workspace-deps` output (expect none).
+
+## T3 — Prove publish/CI unaffected
+
+- **Files:** none (verification).
+- **Do:** confirm `compile-workspace-deps.js` unchanged; re-read each caller
+  (`publish-on-tag.yml`, `.circleci/continue-config.yml`, `new-file-coverage.yml`,
+  `rtk-ci.yml`, `example-*-ci.yml`, `scripts/ci/*.sh`) and confirm each still calls the
+  script directly before its compile step.
+- **Acceptance (IP criterion 5):** every publish/CI compile is preceded by a direct
+  `compile-workspace-deps.js` call; script content unchanged.
+
+## T4 — Measure + docs
+
+- **Files:** build/compile explanation doc (if the root `compile` contract is documented).
+- **Do:** run the metric command best-of-3
+  (`find . -maxdepth 2 -name '*.tsbuildinfo' -not -path '*/node_modules/*' -delete; rm -rf */dist; time bun run compile`);
+  record min. Update the build doc to note reliance on Bun topological ordering (via
+  `update-docs` skill).
+- **Acceptance (IP criteria 4, 6):** cold time recorded with delta vs 305s; `bun run lint`,
+  `bun run api:test`, `bun run ui:test` green.
+- **Verify:** three timings recorded; lint/test exit 0.

package/dist/docs/versioned/next/tasks/create-terreno-app.md

@@ -3,7 +3,7 @@
 Plan: [`docs/implementationPlans/create-terreno-app.md`](../implementationPlans/create-terreno-app.md)
 
 **Status:** Approved — all in-scope tasks Roast PASS. **next: brew**.  
-**Roadmap issue:** https://github.com/FlourishHealth/terreno/issues/1174
+**Roadmap issue:** https://github.com/TerrenoLabs/terreno/issues/1174
 
 ## Instructions for the implementing agent
 

package/dist/docs/versioned/next/tasks/datatable-server-side-filtering.md

@@ -2,7 +2,7 @@
 
 Plan: [`docs/implementationPlans/datatable-server-side-filtering.md`](../implementationPlans/datatable-server-side-filtering.md)  
 **Status:** Pick complete — tasks 1.1–4.1 implemented
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1177
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1177
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/docs-reference-coverage.md

@@ -8,7 +8,7 @@ See: [`docs/implementationPlans/docs-reference-coverage.md`](../implementationPl
 
 - **Do not write documentation from scratch.** The content already exists in `.cursor/rules/*/00-*.mdc` files. Your job is to convert prescriptive agent rules into descriptive public reference prose, verify every claim against source, and remove anything internal.
 - Verify every exported symbol you document by grepping the package's `src/index.ts`. If a rule file documents something that no longer exists, drop it and note the drift in the PR body.
-- READMEs are rendered on npmjs.com where relative links break. Use absolute `https://github.com/flourishhealth/terreno/...` or docs-site URLs in every README.
+- READMEs are rendered on npmjs.com where relative links break. Use absolute `https://github.com/TerrenoLabs/terreno/...` or docs-site URLs in every README.
 - Match the section ordering of `docs/reference/api.md` for new reference pages. Read it first.
 - Run `bun run website:build` before each commit that adds or moves a docs page.
 

package/dist/docs/versioned/next/tasks/examples-demo-coverage.md

@@ -210,7 +210,7 @@ Packages below 95%: none (all published packages meet the 95% function and line
   - Depends on: Task 5.1
   - Acceptance: every gap that a launch document depends on is filled with a working example; remaining gaps are recorded with issue links; both example apps still pass their CI.
 
-Recorded gaps (2026-09-11): AI structured output → [#1010](https://github.com/FlourishHealth/terreno/issues/1010); generic background jobs → [#1188](https://github.com/FlourishHealth/terreno/issues/1188). No launch tutorial currently depends on either.
+Recorded gaps (2026-09-11): AI structured output → [#1010](https://github.com/TerrenoLabs/terreno/issues/1010); generic background jobs → [#1188](https://github.com/TerrenoLabs/terreno/issues/1188). No launch tutorial currently depends on either.
 
 - [x] **Task 5.3**: Link the matrix into the contribution process
   - Description: Add a note to `CONTRIBUTING.md` and the PR template: when adding a framework capability, add or update the example that exercises it and update `docs/explanation/example-coverage.md`. This is what keeps the matrix true. Link the matrix from `docs/explanation/README.md`.

package/dist/docs/versioned/next/tasks/framework-audit-log.md

@@ -1,7 +1,7 @@
 # Tasks: Framework-level audit log
 
 IP: [framework-audit-log.md](../implementationPlans/framework-audit-log.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1186
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1186
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/inbound-webhooks.md

@@ -1,7 +1,7 @@
 # Tasks: Inbound webhook framework
 
 IP: [inbound-webhooks.md](../implementationPlans/inbound-webhooks.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1172
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1172
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/job-queues.md

@@ -1,7 +1,7 @@
 # Tasks: Durable background jobs
 
 IP: [job-queues.md](../implementationPlans/job-queues.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1188
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1188
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/mcp-doc-sync-off-compile-path.md

@@ -0,0 +1,30 @@
+# Tasks: Move mcp-server doc-sync off the compile path
+
+IP: [mcp-doc-sync-off-compile-path.md](../implementationPlans/mcp-doc-sync-off-compile-path.md)
+Slice: mcp-doc-sync-off-compile-path · Scan goal: reduce-cold-compile-time (305s → 214s)
+
+## T0 — Baseline snapshot (done)
+
+- Built current mcp `compile` (full pipeline) and snapshotted `dist/docs`
+  (3224 versioned + 6 guidelines + ui-types) for parity diff.
+
+## T1 — Split compile/build + bundle docs at publish (done)
+
+- `mcp-server/package.json`: `compile` → `NODE_OPTIONS=--max-old-space-size=8192 tsc`;
+  `build` keeps sync+tsc+cp, redundant `NODE_OPTIONS` dropped from the copy steps.
+- Added `prepublishOnly: "bun run build"` so **both** publish paths (GitHub Actions and
+  CircleCI `scripts/ci/publish-package.sh`) bundle docs via npm's lifecycle. No workflow
+  edit needed. (Fixes Bugbot: CircleCI publish also runs the tsc-only `compile`.)
+
+## T2 — Verify (done)
+
+- mcp `compile`: tsc only, 0 "Synced docs" lines, no `dist/docs`, `dist/index.js` present. ✅
+- mcp `build`: `dist/docs` byte-identical to baseline (`diff -r` → IDENTICAL). ✅
+- Combined A+B integration (B stacked temporarily): full `bun run compile` ×3 →
+  exit 0, 0 errors, 0 mcp sync during compile. Best cold **186s vs 305s baseline (−39%)**. ✅
+- publish-mcp job runs `bun run build`. ✅
+
+## T3 — Docs (done)
+
+- Updated `.claude/rules/mcp-server/00-mcp-server.md`: `compile` = type-check,
+  `build` = shippable artifact with docs.

package/dist/docs/versioned/next/tasks/migrate-cicd-to-circleci.md

@@ -15,7 +15,7 @@ Research: [`docs/implementationPlans/migrate-cicd-to-circleci-research.md`](../i
 ## Phase 0: Prerequisites
 
 - [ ] **Task 0.1**: Confirm CircleCI org/plan and project link
-  - Description: Verify the requester-created CircleCI organization, plan, and that `FlourishHealth/terreno` is connected with permission to report GitHub checks. Enable **dynamic config** / setup workflows for the project. Record org slug + project slug in the IP header or how-to doc.
+  - Description: Verify the requester-created CircleCI organization, plan, and that `TerrenoLabs/terreno` is connected with permission to report GitHub checks. Enable **dynamic config** / setup workflows for the project. Record org slug + project slug in the IP header or how-to doc.
   - Files: `docs/how-to/circleci.md` (stub OK)
   - Depends on: none (human prerequisite)
   - Acceptance: a manual “hello” pipeline can be triggered from the project; dynamic config is enabled.
@@ -178,7 +178,7 @@ Research: [`docs/implementationPlans/migrate-cicd-to-circleci-research.md`](../i
   - Description: Port `maestro-e2e` (Chrome + xvfb against example-frontend, optional demo smokes) to CircleCI `node22_browsers_mongo_rs`. Appium Android remains separate.
   - Files: `.circleci/continue-config.yml`, `.circleci/config.yml`, `.github/workflows/maestro-e2e.yml`
   - Depends on: Task 5.2
-  - Acceptance: CircleCI `maestro-e2e` is the writer; GHA workflow is `on: []`.
+  - Acceptance: CircleCI `maestro-e2e` is the writer; GHA workflow is `push.branches-ignore: ["**"]`.
 
 - [ ] **Task 8.2**: Port Appium iOS on CircleCI macOS
   - Description: Select macOS resource class/image with Working WDA; port iOS job; compare flake rate to GHA `macos-15`.
@@ -207,7 +207,7 @@ Research: [`docs/implementationPlans/migrate-cicd-to-circleci-research.md`](../i
   - Acceptance: each former workflow’s acceptance behavior has an owner on CircleCI; GHA files deleted.
 
 - [x] **Task 9.3a**: Port architectural PR review
-  - Description: CircleCI `architectural-pr-review` runs `.github/scripts/architectural-pr-review.ts` after checking out `origin/master`. Skips forks and missing `CURSOR_API_KEY` / `GITHUB_TOKEN`. GHA workflow is `on: []`.
+  - Description: CircleCI `architectural-pr-review` runs `.github/scripts/architectural-pr-review.ts` after checking out `origin/master`. Skips forks and missing `CURSOR_API_KEY` / `GITHUB_TOKEN`. GHA workflow is `push.branches-ignore: ["**"]`.
   - Files: `.circleci/continue-config.yml`, `.github/workflows/architectural-pr-review.yml`, `.github/scripts/architectural-pr-review.ts`
   - Depends on: Task 0.2
   - Acceptance: CircleCI job is the writer; Cursor Approval Agent stays on GitHub.

package/dist/docs/versioned/next/tasks/mongodb-migrations-tooling.md

@@ -1,7 +1,7 @@
 # Tasks: MongoDB migrations tooling
 
 IP: [mongodb-migrations-tooling.md](../implementationPlans/mongodb-migrations-tooling.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1189
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1189
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/notification-center.md

@@ -1,7 +1,7 @@
 # Tasks: In-app notification center
 
 IP: [notification-center.md](../implementationPlans/notification-center.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1179
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1179
 
 **Feature profile:** false (full IP)
 

package/dist/docs/versioned/next/tasks/rate-limiting.md

@@ -1,7 +1,7 @@
 # Tasks: API rate limiting
 
 IP: [rate-limiting.md](../implementationPlans/rate-limiting.md)  
-**Closes:** https://github.com/FlourishHealth/terreno/issues/1187
+**Closes:** https://github.com/TerrenoLabs/terreno/issues/1187
 
 **Feature profile:** false (full IP)