fix: follow-ups to #290 and #292 (subject-set wording, re-run contradictions, baseline cost)

CHANGELOG.md

@@ -17,6 +17,8 @@ Sections dated before 2026-09-19 predate the cycle and stay as they are.
 
 ## Unreleased
 
+- fix(testing-shape, module-review, dispatching): **follow-ups to #290 and #292 — subject set reaches embedders, reviewers never re-run, dispatch guidance is its own small baseline file.** A UI-only change now re-runs the journeys of every page it reaches (pages embedding a changed snippet or layout, in any module), not "no other module's journeys"; `module-review.md`'s Prove row and "nothing grandfathers" line no longer say "re-run after any change" (implementer re-runs the subject set once, reviewer verifies the artifact, per `testing-shape.md` §3); the #274 cost anecdote appears once in `testing-shape.md` and is one clause in intake Q13, instead of told three times in full; `module-review.md` states the PROOF-OF-LOOK addenda rule and the final LOOK's `VALID AT:`/per-page proof/Stage-6 requirements that #321's checker enforces. `agent-roles.md`'s Dispatching section moved to `skills/dispatching.md` (981 words, with a note that forks/resumes/per-call models are Claude Code concepts — elsewhere, one fresh session per task); the lead loads that at P–6 and `agent-roles.md` returns to its pre-#292 setup row (P,4,5). Lead load at stages 1/2/3/6: −2,059 words each; baseline 76,259 → 76,191 ([#274](https://github.com/MendixMau/mxcli-project-toolkit/issues/274), [#288](https://github.com/MendixMau/mxcli-project-toolkit/issues/288)) — MendixMau
+- chore(context): **baseline word budget freed — 79,997 → 76,259 words (−3,738) without dropping a rule.** The always-on tier sat 3 words under its 80,000 cap, so three open PRs could not all merge. Long incident narratives in `conversion-runbook.md`, `source-triage.md` and `interview-protocol.md` became one-line failure examples pointing at new on-demand `process/baseline-incident-history.md`, where the original text is kept verbatim per section (R1–R20, AR1, ST1–ST4, IP1–IP3); the runbook's copy of the open-questions status table now defers to `interview-protocol.md` §4; `agent-roles.md`'s six template shapes (setup-only, ~2,300 words) moved verbatim to `agents/README.md`, which the installers never copy (explicit `MXTK_AGENTS` list). Every STOP, rule, denominator and gate cell stays in place; files touched by open PRs were left alone. — 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

@@ -764,7 +764,8 @@ Then filter by Role(s). A helper dispatched with a role stub (ba, architect, mdl
 | Putting a question TO the user — any gate, any stage: ask in chat not in a file, two named options plus your recommendation, one batch per gate then end the turn | `skills/interview-protocol.md` | P,0,1,2,3,4,5,6,7 | lead, ba, architect |
 | The user gives feedback or asks for a change mid-project — "change X", "it also needs Y", a remark on a screenshot or the running app, a stakeholder's ask relayed in chat: log a CR row in PROJECT.md ## Change log BEFORE any edit, size it, then edit the highest artifact it alters (BRD, decision, architecture, wireframe), never the MDL first | `skills/change-intake.md` | 2,3,4,5,6,7 | lead, ba |
 | Any stage transition — the 2+1 format every CAC uses, and the one-register rule (answers land in PROJECT.md, never in a separate state file). The seven CACs themselves are routed per stage in the situational table | `skills/checkpoints/checkpoint-template.md` | 0,1,2,3,4,6,7 | lead, ba, architect |
-| Starting a subagent, sending it a fix round, or reading an agent-stop notice; or setting up the project's subagents — read its Dispatching section per dispatch, the rest once at setup | `skills/agent-roles.md` | P,1,2,3,4,5,6 | lead |
+| Setting up or completing a project's dev-process subagents — once, at project start, not "on demand" | `skills/agent-roles.md` | P,4,5 | lead |
+| Starting a subagent, sending it a fix round, or reading an agent-stop notice — name the model, fresh not fork, fresh fixer not resume, coalesce notices | `skills/dispatching.md` | P,1,2,3,4,5,6 | lead |
 | Deciding whether to extract at all, before any BRD gets generated | `skills/source-triage.md` | 0 | ba |
 | Taking in a new source — before generating anything from it. Grades what the source can support; nothing else in this toolkit reads a source | `bin/source-sufficiency.sh` | 0,1 | ba |
 | Closing Stage 1, or adding files to a source folder — every inventoried file must name the artifact that consumed it (text AND embedded diagrams), or carry a waiver; blocks Stages 1–2 until it does | `bin/source-ledger.sh` | 0,1,2 | ba |

ROUTING.md

@@ -47,7 +47,8 @@ picks the row up. That is the whole procedure — there is no second list to rem
 | CAC-4, after rearchitect sign-off and before any design artifact — branding and UI direction. Opens with a brainstorm | `skills/checkpoints/checkpoint-design.md` | architect | 3 | ondemand |
 | CAC-5, after design sign-off and before the build plan — build order and slice boundaries. Opens with a brainstorm | `skills/checkpoints/checkpoint-build.md` | architect,mdl | 4 | ondemand |
 | CAC-6, after Stage 6 passes and before any cutover step — migration mode only, and a hard gate: every answer lands CONFIRMED, no ASSUMED defaults | `skills/checkpoints/checkpoint-cutover.md` | gate,test | 6,7 | ondemand |
-| Starting a subagent, sending it a fix round, or reading an agent-stop notice; or setting up the project's subagents — read its Dispatching section per dispatch, the rest once at setup | `skills/agent-roles.md` | lead | P,1,2,3,4,5,6 | baseline |
+| Setting up or completing a project's dev-process subagents — once, at project start, not "on demand" | `skills/agent-roles.md` | lead | P,4,5 | baseline |
+| Starting a subagent, sending it a fix round, or reading an agent-stop notice — name the model, fresh not fork, fresh fixer not resume, coalesce notices | `skills/dispatching.md` | lead | P,1,2,3,4,5,6 | baseline |
 | Stage 0 sign-off when the inventory is at or under 1 module / 8 screens / 25 use cases, or the user says the app is small — declare the tier, then apply its per-stage caps and the three artifact waivers | `skills/small-project-tier.md` | ba,architect | 0,1,2,3,4 | ondemand |
 | Deciding who answers a question — before putting any batch to the user. gap/conflict/choice/user-only is what keeps a gate batch at four questions instead of 127 | `bin/question-kinds.sh` | ba | 1,2,3 | baseline |
 | Generating a new project's CLAUDE.md — baseline routing plus project-specific facts | `skills/bootstrap-project.md` | ba | P | ondemand |