Promote 2026-09-29 inbox notes: exec.sh refusal rule, UI fix rounds, transcript usage, autocompact advisory

CHANGELOG.md

@@ -17,6 +17,9 @@ Sections dated before 2026-09-19 predate the cycle and stay as they are.
 
 ## Unreleased
 
+- learn(iterative-build-loop, upstream-feedback): **every model write goes through `exec.sh`; on a refusal the move is commit (`--commit-first`), wait (exec lock) or stop and report — never `FORCE_EXEC` on the agent's own initiative; MCP writes stay allowed and are accepted as unrecorded, recorded by an `MCP:` commit; the gate checks the saved file (save, then check); no batched-gate mode exists, so do not imitate one with overrides; Studio Pro MCP server asks get a third destination in `upstream-feedback.md` with their evidence rows** ([#69](https://github.com/MendixMau/mxcli-project-toolkit/issues/69), [#171](https://github.com/MendixMau/mxcli-project-toolkit/issues/171), [#173](https://github.com/MendixMau/mxcli-project-toolkit/issues/173)). The phase-close check is coverage, N of N: every model commit is covered by an applied exec row, a `--commit-first` commit or an `MCP:` commit (the pre-exec sequence now commits as `MCP: …`); the refusal table includes PRE-EXISTING errors and ends with "read the printed remedy"; the exec lock refuses rather than queues, and the text says so. Field evidence: a 2026-09-29 two-build comparison (229 errors hidden by a stale disk copy; per-script apply 22–30 s growing to 55–65 s). Routing triggers for the build loop, page patterns and measured claims now name the refusal, the UI round and transcript token counts (stubs re-rendered; baseline unchanged at 79,997 words) — MendixMau
+- learn(learned-page-patterns, measured-claims): **UI fix rounds: SCSS first, apply directly through `exec.sh`, re-look only changed pages (N of N), and check four regressions before applying (N1 page-column on the layout root, N2 `.d-none` losing to a `display` class, N3 Data Grid 2 `autoFit` columns at 32 px, N4 focus ring); transcript usage is read from each message's LAST record** ([#170](https://github.com/MendixMau/mxcli-project-toolkit/issues/170), [#171](https://github.com/MendixMau/mxcli-project-toolkit/issues/171)). The round took 1,507 calls against 515 for a comparable build; re-measured on Claude Code 2.1.296 transcripts, sub-agent output read from the first record was up to 7x low while the main loop's was exact — MendixMau
+- new(init-project): **the scaffold prints an auto-compact advisory: a warning when `CLAUDE_CODE_AUTO_COMPACT_WINDOW` or any `autoCompactWindow` in the project or user settings is above 250k, an "unparseable" line for a value it cannot read, otherwise a `/autocompact 250k` recommendation; it writes no setting and never fails the scaffold** ([#167](https://github.com/MendixMau/mxcli-project-toolkit/issues/167)). A lead session at a 1M window averaged 363–466k context per call. Fixture `tests/wave2/test-init-autocompact.sh` (15 checks; fails on master) — MendixMau
 - learn(testing-shape, test-result-audit): **three test-strength rules, each with a denominator: refusals probed on the server, every undeclared 4xx a failure, one clean-database final run** ([#172](https://github.com/MendixMau/mxcli-project-toolkit/issues/172)). A field comparison of two builds against one frozen spec found the toolkit build checked permission refusals only in the client (a hidden menu item), logged unexpected 400/422 responses as information, and reached 20/20 by stitching three runs. `testing-shape.md` §7 gains three report lines (`N of N restricted pairs probed server-side`, `undeclared 4xx/5xx: 0 of M requests — each fails its step`, `one clean-DB run <id>: N of N scenarios incl. timers, not stitched`); `test-result-audit.md` gains check 2e, which carries the reasons and the field evidence and asks for those numbers. Both baseline skills stay inside the 80,000-word baseline budget (79,997): the reasons live in the on-demand audit skill, and two redundant sentences in `testing-shape.md` §4's verdict paragraphs were folded (the empty-set corollary is stated in the Trace row, `harness-architecture.md` and `report-schema.md`). The backing inbox note is trimmed to its unpromoted "Keep" items. — MendixMau
 - learn(learned-detection-gaps, module-review): **two UI defects every model gate passes: a lazy-loaded combo box filter hides its preset, and a theme that pins row heights by position lets the next row cover an inserted widget** ([#299](https://github.com/MendixMau/mxcli-project-toolkit/issues/299)). Both passed check, mxbuild, lint and the scripted checks on a 127-module production app and were found only in the running app. Two register rows with the first oracle and the pre-build check (pluggable properties that change *when* data loads; grep the theme for fixed `height` on positional row classes before an insert), and two LOOK lines in `module-review.md` §4c with denominators (`N of N filters show their preset on first load`, `N of N widgets inserted into an existing container fully visible`). — jopterhorst
 - fix(app-restart, test-stack-up): **a port held by another process before `run --local` is named and reported as "instrument blocked" (exit 4), never killed** ([#262](https://github.com/MendixMau/mxcli-project-toolkit/issues/262)). Studio Pro had started its own runtime on the app/admin ports a restart needed; `mxcli run --local` then failed with "Authentication failed" (the admin API answering was Studio Pro's), the next journey ran against Studio Pro's runtime, and four more runs were NOT RUN. `app-restart.sh` now checks the app port and the admin port (record's `adminPort`, else app + 10) before stopping anything — a holder outside the recorded process tree exits 4 with its name, pid and parent — and again after the stop, where anything still holding a port is by definition not ours. `test-stack-up.sh` runs the same check wherever it would tell you to start `run --local` (`--no-docker`, no Docker daemon, `--check` with a leftover record), on the record's ports, `RUN_LOCAL_APP_PORT`/`RUN_LOCAL_ADMIN_PORT`, or mxcli's defaults 8080/8090. Holder lookup: lsof, then ss, then netstat + tasklist on Windows; unnamed holders still block when nothing of ours is running. Fixtures: `test-app-restart.sh` +2 cases (14 assertions, all fail on the old script), `test-stack-up-local.sh` T7 (4 of 6 fail on the old script). — AlvinnYeo

README.md

@@ -525,7 +525,7 @@ Every mxcli project has a `.ai-context/skills/` directory (bundled by `mxcli ini
 | Changing an EXISTING Mendix app — adding a feature, altering a flow, restructuring a module — when it has no BRDs, no architecture doc and no wireframes: the knowledge base comes from the live model (Path D), stages 2–4 run over what changes plus what it touches, nothing more, and the Track B regression baseline is the precondition; audit-only stays in existing-app-assurance | `skills/existing-app-change.md` |
 | Setting up or wiring a COMPANY BRAIN — the private tier between the toolkit and a project for own skills, conventions, lint rules, MDL snippets and approved MPKs; and deciding whether something goes to the toolkit, the company brain or docs/brain/ | `skills/company-brain.md` |
 | Cutover and retrospective — promoting proven patterns back into the toolkit | `skills/close-the-loop.md` |
-| Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence — a claim not in the register may not be cited | `skills/measured-claims.md` |
+| Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence, or counting tokens from transcripts — a claim not in the register may not be cited | `skills/measured-claims.md` |
 | Any review pass that runs more than once — module-review, coherence, monkey, wiring-sweep: findings accumulate across runs, a per-run report cannot show a trend | `skills/improvement-register.md` |
 
 **Source — reading a legacy system (migration entry mode)**
@@ -587,7 +587,7 @@ Every mxcli project has a `.ai-context/skills/` directory (bundled by `mxcli ini
 | Task | Skill to load |
 |---|---|
 | Dispatching a build step to a helper — generate its one-file pack (the brief's Build steps row, mxcli brain brief, live DESCRIBEs, example, folder) and hand over the path instead of a reading list. Measured: 25-38% fewer tool calls and 30-41% less time, same quality; the token saving is not proven | `project-bin/context-pack.sh` |
-| Building a module with mxcli — verified, iterative, coverage-checklist gated | `skills/iterative-build-loop.md` |
+| An exec.sh guard refused a write (not a permission prompt), or building a module with mxcli — verified, iterative, coverage-checklist gated; on a refusal: commit, wait or stop, never FORCE_EXEC | `skills/iterative-build-loop.md` |
 | After marking a module done, or any time "how much is built vs proven" is asked — renders build-plan.html from done- prefixes and verify-module.sh/improvement-register.md, kept as two honestly separate views; --json writes architecture/build-plan.json parsed from build-plan.md's Phase headings (a plan with no Phase headings gets no file) | `project-bin/build-plan-status.sh` |
 | Turning a client-derived Mendix app into a clean, shareable demo with zero client fingerprint — branding, data, custom widgets | `skills/anonymize-client-app-for-demo.md` |
 | Handing a headless-built model to a person — opening it in Studio Pro, a free sandbox, or a colleague's machine: the model travels, the demo data and runtime config (keys, an agent's bound model) do not, and each needs its own re-establish step | `skills/handoff-to-studio-pro.md` |
@@ -610,7 +610,7 @@ Every mxcli project has a `.ai-context/skills/` directory (bundled by `mxcli ini
 
 | Task | Skill to load |
 |---|---|
-| Building and auditing Mendix pages — widget patterns, datasource shapes | `skills/learned-page-patterns.md` |
+| Building and auditing Mendix pages, or starting a UI fix round across many pages — widget patterns, datasource shapes; for a round: SCSS first, four regressions to check | `skills/learned-page-patterns.md` |
 | Generating a whole page tree in one script — the structure patterns that survive it | `skills/oneshot-page-structure-patterns.md` |
 | Building or auditing a collapsible sidebar nav — Atlas Core's collapsed state needs icons assigned per menu item or it silently clips label text | `skills/learned-sidebar-collapse-icons.md` |
 | Building or altering any data grid — native DATAGRID vs pluggable DG2 decision rule, the ALTER PAGE INSERT corruption, sort-by and filter-binding traps | `skills/learned-dg2-patterns.md` |

ROUTING.md

@@ -55,7 +55,7 @@ picks the row up. That is the whole procedure — there is no second list to rem
 | Changing an EXISTING Mendix app — adding a feature, altering a flow, restructuring a module — when it has no BRDs, no architecture doc and no wireframes: the knowledge base comes from the live model (Path D), stages 2–4 run over what changes plus what it touches, nothing more, and the Track B regression baseline is the precondition; audit-only stays in existing-app-assurance | `skills/existing-app-change.md` | ba,architect | P,0 | ondemand |
 | Setting up or wiring a COMPANY BRAIN — the private tier between the toolkit and a project for own skills, conventions, lint rules, MDL snippets and approved MPKs; and deciding whether something goes to the toolkit, the company brain or docs/brain/ | `skills/company-brain.md` | all | - | ondemand |
 | Cutover and retrospective — promoting proven patterns back into the toolkit | `skills/close-the-loop.md` | all | 7 | ondemand |
-| Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence — a claim not in the register may not be cited | `skills/measured-claims.md` | all | - | ondemand |
+| Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence, or counting tokens from transcripts — a claim not in the register may not be cited | `skills/measured-claims.md` | all | - | ondemand |
 | Any review pass that runs more than once — module-review, coherence, monkey, wiring-sweep: findings accumulate across runs, a per-run report cannot show a trend | `skills/improvement-register.md` | mdl,gate,test,review | 5,6 | ondemand |
 | The first command of every session, and any time someone asks "where are we" or "what next" — one screen: stage, done/overdue, the ONE next action, from the instruments, never from memory | `bin/status.sh` | all | P,0,1,2,3,4,5,6,7 | baseline |
 | Before obeying any learned-* STOP or workaround that costs a detour — probe the binary you actually have, then stamp the verdict back into the rule | `skills/retesting-learned-rules.md` | all | - | baseline |
@@ -136,7 +136,7 @@ picks the row up. That is the whole procedure — there is no second list to rem
 | Reviewing any module before calling it done — the ONE pass: build, gate, prove, LOOK (is it logical, does it look right, does it match our design, over every page not just the tested ones), confirm with the denominator stated | `skills/module-review.md` | mdl,review,test | 5,6 | baseline |
 | Before any mxcli exec / exec.sh / --mcp write — ask or run? the knob decides | `bin/exec-approval.sh` | mdl,gate | 5,6 | baseline |
 | Dispatching a build step to a helper — generate its one-file pack (the brief's Build steps row, mxcli brain brief, live DESCRIBEs, example, folder) and hand over the path instead of a reading list. Measured: 25-38% fewer tool calls and 30-41% less time, same quality; the token saving is not proven | `project-bin/context-pack.sh` | mdl | 5 | ondemand |
-| Building a module with mxcli — verified, iterative, coverage-checklist gated | `skills/iterative-build-loop.md` | mdl,gate | 5 | ondemand |
+| An exec.sh guard refused a write (not a permission prompt), or building a module with mxcli — verified, iterative, coverage-checklist gated; on a refusal: commit, wait or stop, never FORCE_EXEC | `skills/iterative-build-loop.md` | mdl,gate | 5 | ondemand |
 | After marking a module done, or any time "how much is built vs proven" is asked — renders build-plan.html from done- prefixes and verify-module.sh/improvement-register.md, kept as two honestly separate views; --json writes architecture/build-plan.json parsed from build-plan.md's Phase headings (a plan with no Phase headings gets no file) | `project-bin/build-plan-status.sh` | architect,gate,review | 4,5,6 | ondemand |
 | Turning a client-derived Mendix app into a clean, shareable demo with zero client fingerprint — branding, data, custom widgets | `skills/anonymize-client-app-for-demo.md` | mdl,review | 6 | ondemand |
 | Handing a headless-built model to a person — opening it in Studio Pro, a free sandbox, or a colleague's machine: the model travels, the demo data and runtime config (keys, an agent's bound model) do not, and each needs its own re-establish step | `skills/handoff-to-studio-pro.md` | test,review | 6,7 | ondemand |
@@ -164,7 +164,7 @@ picks the row up. That is the whole procedure — there is no second list to rem
 
 | Always relevant for | Load this | Agent(s) | Stage(s) | Tier |
 |---|---|---|---|---|
-| Building and auditing Mendix pages — widget patterns, datasource shapes | `skills/learned-page-patterns.md` | mdl | 5 | ondemand |
+| Building and auditing Mendix pages, or starting a UI fix round across many pages — widget patterns, datasource shapes; for a round: SCSS first, four regressions to check | `skills/learned-page-patterns.md` | mdl | 5 | ondemand |
 | Generating a whole page tree in one script — the structure patterns that survive it | `skills/oneshot-page-structure-patterns.md` | mdl | 5 | ondemand |
 | Building or auditing a collapsible sidebar nav — Atlas Core's collapsed state needs icons assigned per menu item or it silently clips label text | `skills/learned-sidebar-collapse-icons.md` | mdl | 5 | ondemand |
 | Building or altering any data grid — native DATAGRID vs pluggable DG2 decision rule, the ALTER PAGE INSERT corruption, sort-by and filter-binding traps | `skills/learned-dg2-patterns.md` | mdl | 5 | ondemand |

agents/architect-agent.md

@@ -71,7 +71,7 @@ This table is your whole list. The baseline table in the project's CLAUDE.local.
 | `skills/mendix-agents.md` | Building a Mendix AI agent |
 | `skills/company-brain.md` | Setting up or wiring a COMPANY BRAIN |
 | `skills/close-the-loop.md` | Cutover and retrospective |
-| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence |
+| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence, or counting tokens from transcripts |
 | `skills/agent-permission-friction.md` | Any refused, denied or blocked command |
 | `project-bin/build-plan-status.sh` | After marking a module done, or any time "how much is built vs proven" is asked |
 | `skills/workflow-structure-rules.md` | Designing or reviewing a Workflow's SHAPE before or after the MDL |

agents/ba-agent.md

@@ -87,7 +87,7 @@ This table is your whole list. The baseline table in the project's CLAUDE.local.
 | `skills/brd-validation.md` | Validating BRDs against the code and document KB |
 | `skills/company-brain.md` | Setting up or wiring a COMPANY BRAIN |
 | `skills/close-the-loop.md` | Cutover and retrospective |
-| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence |
+| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence, or counting tokens from transcripts |
 | `skills/agent-permission-friction.md` | Any refused, denied or blocked command |
 | `skills/mendix-epics-api.md` | Working with the Mendix Epics board programmatically |
 | `skills/upstream-feedback.md` | About to open an issue, PR or discussion against mxcli or the toolkit |

agents/gate-agent.md

@@ -56,14 +56,14 @@ This table is your whole list. The baseline table in the project's CLAUDE.local.
 | `skills/cloud-dev-environment.md` | Setting up or resuming an mxcli project in a cloud/ephemeral container |
 | `skills/coverage-ledger.md` | Building the Stage 4 coverage ledger |
 | `bin/coverage-check.sh` | Checking a coverage ledger against its BRD |
-| `skills/iterative-build-loop.md` | Building a module with mxcli |
+| `skills/iterative-build-loop.md` | An exec.sh guard refused a write (not a permission prompt), or building a module with mxcli |
 | `project-bin/conformance-check.sh` | Running the ledger rung alone |
 | `project-bin/graph-sweep.sh` | Running the wiring rung alone |
 | `project-bin/test-stack-up.sh` | Before any runtime test |
 | `skills/company-brain.md` | Setting up or wiring a COMPANY BRAIN |
 | `skills/close-the-loop.md` | Cutover and retrospective |
 | `skills/report-schema.md` | Writing or reading docs/report.json |
-| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence |
+| `skills/measured-claims.md` | Before citing ANY behavioural claim about the harness, the Mendix runtime or a test tool as evidence, or counting tokens from transcripts |
 | `skills/agent-permission-friction.md` | Any refused, denied or blocked command |
 | `project-bin/coherence-cadence.sh` | After every module's CONFIRM stage |
 | `project-bin/build-plan-status.sh` | After marking a module done, or any time "how much is built vs proven" is asked |