humanish 0.53.0 → 0.55.0

package/dist/tui-contract.d.ts

@@ -51,6 +51,8 @@ export interface TuiCapabilities {
     openObserver(cwd: string, observerPath: string): Promise<TuiActionResult>;
     /** Stop the sandboxes an interrupted run left behind, keeping its evidence. */
     reclaimRun(cwd: string, runId: string): Promise<ReclaimResult>;
+    /** End a run that is still going. Stops the PROCESS; `reclaimRun` stops what it provisioned. */
+    stopRun(cwd: string, runId: string): Promise<TuiActionResult>;
 }
 export interface TuiOptions {
     /** The project the surface is reading. Already resolved by the CLI. */

package/dist/tui-contract.js.map

@@ -1 +1 @@
-{"version":3,"file":"tui-contract.js","sourceRoot":"","sources":["../src/tui-contract.ts"],"names":[],"mappings":"AAAA,2DAA2D;AAC3D,EAAE;AACF,6FAA6F;AAC7F,mGAAmG;AACnG,+EAA+E;AAC/E,EAAE;AACF,gGAAgG;AAChG,+FAA+F;AAC/F,gGAAgG;AAChG,mGAAmG;AACnG,6DAA6D;AAC7D,EAAE;AACF,mGAAmG;AACnG,mBAAmB;AAsFnB,mFAAmF;AACnF,MAAM,CAAC,MAAM,kBAAkB,GAAG,EAAE,CAAC;AAErC;;;GAGG;AACH,MAAM,UAAU,eAAe,CAAC,gBAAwB,OAAO,CAAC,OAAO;IACrE,MAAM,KAAK,GAAG,MAAM,CAAC,QAAQ,CAAC,aAAa,CAAC,OAAO,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,EAAE,EAAE,CAAC,CAAC;IACvF,OAAO,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,IAAI,kBAAkB,CAAC;AAC/D,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,YAAY,CAAC,OAAe;IAC1C,OAAO,IAAI,GAAG,CAAC,cAAc,EAAE,OAAO,CAAC,CAAC;AAC1C,CAAC"}
+{"version":3,"file":"tui-contract.js","sourceRoot":"","sources":["../src/tui-contract.ts"],"names":[],"mappings":"AAAA,2DAA2D;AAC3D,EAAE;AACF,6FAA6F;AAC7F,mGAAmG;AACnG,+EAA+E;AAC/E,EAAE;AACF,gGAAgG;AAChG,+FAA+F;AAC/F,gGAAgG;AAChG,mGAAmG;AACnG,6DAA6D;AAC7D,EAAE;AACF,mGAAmG;AACnG,mBAAmB;AAwFnB,mFAAmF;AACnF,MAAM,CAAC,MAAM,kBAAkB,GAAG,EAAE,CAAC;AAErC;;;GAGG;AACH,MAAM,UAAU,eAAe,CAAC,gBAAwB,OAAO,CAAC,OAAO;IACrE,MAAM,KAAK,GAAG,MAAM,CAAC,QAAQ,CAAC,aAAa,CAAC,OAAO,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,EAAE,EAAE,CAAC,CAAC;IACvF,OAAO,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,IAAI,kBAAkB,CAAC;AAC/D,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,YAAY,CAAC,OAAe;IAC1C,OAAO,IAAI,GAAG,CAAC,cAAc,EAAE,OAAO,CAAC,CAAC;AAC1C,CAAC"}

package/docs/contracts/schemas.md

@@ -3,7 +3,7 @@
 Date: 2026-06-02 (current-state note updated 2026-07-14)
 
 Status: reference map for the major contracts shipped through source version
-`0.53.0`; it is not an exhaustive inventory of command/result envelopes. Exported types,
+`0.55.0`; it is not an exhaustive inventory of command/result envelopes. Exported types,
 schema constants, parsers, and validators in `src/` are authoritative. Rows
 marked "reserved" name layering intent only — no code emits or validates them
 yet. Do not emit a reserved schema.
@@ -46,6 +46,7 @@ workflow without leaking private upstream truth into core.
 | Pricing (operator-editable rates) | `humanish.pricing.v1` (`src/pricing.ts`; dated per-model + E2B desktop rates) | see Run Cost Summary And Estimated Actor Cost below |
 | Run cost summary | `humanish.run-cost-summary.v1` (additive `RunBundle.cost`; estimate, never a charge) | see Run Cost Summary And Estimated Actor Cost below |
 | Estimated actor cost | `humanish.actor-estimated-cost.v1` (additive `ActorTrace.estimatedCost`) | see Run Cost Summary And Estimated Actor Cost below |
+| Model settings | `humanish.model-settings.v1` (additive `ActorTrace.modelSettings`; the reasoning effort the request carried) | see Actor Trace below |
 | Affordance use | `humanish.affordance-use.v1` (additive `ActorTrace.affordanceUse`; per-class counts of the routes an actor took) | see Affordance Use below |
 | Adapter score | `humanish.adapter-score.v1` (`RunBundle.adapterScore`; namespaced; route-specific acceptance semantics) | see Product-Adapter Extension Seam below |
 | Adapter artifact | `humanish.adapter-artifact.v1` (`RunBundle.adapterArtifacts[]`; namespaced; local relative proof references) | see Product-Adapter Extension Seam below |
@@ -741,6 +742,16 @@ Core-owned fields:
   click-like `ui_action` items, structured `coord` (`x`/`y`) — both additive
   (#441): absent on older bundles and non-stamping producers, and absence means
   "timing/position unrecorded", never zero
+- optional `modelSettings` (`humanish.model-settings.v1`, #497): HOW the model was
+  asked to run, alongside `ids.model` which says which model it was. Today one
+  field, `reasoningEffort`. It records the value the request ACTUALLY carried,
+  including the provider default when a lab declared nothing — the resolved
+  value is what produced the trace, so reporting it is honest where reporting
+  "unset" would not be. Absent when a provider declares no settings and on every
+  pre-existing bundle; tolerated by verify. It exists because effort was
+  unreachable from a lab, which made every run take the default silently, and a
+  trace that cannot say what effort produced it cannot be compared with one that
+  can
 - optional `affordanceUse` (`humanish.affordance-use.v1`): which KIND of route this
   actor took (see Affordance Use below)
 - optional `estimatedCost` (`humanish.actor-estimated-cost.v1`): a token-derived

package/docs/goals/current.md

@@ -29,7 +29,7 @@ study completed, reproduced, and produced a real accessibility finding via a
 keyboard-first participant
 ([docs/goals/email-gated-signup/receipts/](email-gated-signup/receipts/)).
 
-## Current Program Truth (source `0.53.0`)
+## Current Program Truth (source `0.55.0`)
 
 The package source and repository implementation in this tree agree on these
 points:
@@ -39,14 +39,14 @@ The immutable 2026-06-10 proof-roadmap packet is paired with a
 
 | Surface | Shipped | Still unproven or unbuilt |
 | --- | --- | --- |
-| Actor execution | Six first-party registry descriptors; computer-use, scripted-browser, and terminal-product dispatch paths | Public out-of-tree actor registration and conformance certification |
+| Actor execution | Six first-party registry descriptors; computer-use, scripted-browser, and terminal-product dispatch paths. `0.55.0` makes reasoning effort declarable per actor and per lane and records the resolved value on the trace (`humanish.model-settings.v1`) — it had been reachable in the provider and unreachable from a lab, so every prior run silently took the provider default | Public out-of-tree actor registration and conformance certification |
 | Persona scale | Bounded per-lane-world fan-out, including differentiated lanes and roster expansion; kept deterministic and live receipts | A completed first-party deletion branch that replaces a bespoke generic harness |
 | Shared state | Sequential and concurrent single-origin shared-world execution; sequential has deterministic proof, concurrent has deterministic and kept live proof | Multi-origin shared-world runtime/schema support; real-adopter deletion proof |
 | Subject sources/routes | Six declared sources: `this-repo`, `clone`, `app-url`, `local-app`, `terminal-product`, and `local-tree`; support is route-specific and `this-repo` remains dry-run-only | One centralized run/resource lifecycle boundary across all routes |
 | Public proof | A legible four-persona Observer hero from a verified real public-application study (commit-pinned drawDB) shipped in the npm payload (`0.16.0`) | Coverage beyond a single studied subject; the stratified breadth panel remains unbuilt |
 | OSS meta-lab | Dry-run contract and separate disposable smoke harness | Live meta-lab execution; disabled until repository instructions and actor credentials have an isolated boundary |
 | Observer serving | `watch`/`observe` loopback servers plus `serve` — the run-library surface with loopback default, capability-link exposure, `share_ready`-gated open mode, and optional operator-run tunnel; streams never served remotely | A remote live-stream (`--live-streams`) design; a persistent capability-link store |
-| Stakeholder terminal surface | `humanish tui` (`0.50.0`, redesigned to the reviewed spec in `0.51.0`): labs -> lab -> run, arrow-key navigation, and starting a dry or live run from the lab screen. The run is DETACHED and outlives the terminal — live-proven by killing the terminal 42s into a real run that then ran on for ~3.5 minutes and finished `pass` at $0.639751. The run screen leads with the participant and their recorded thinking, live-proven mid-flight against a real computer-use run. Ships as one bundled file loaded on demand; refuses a non-interactive stdin/stdout naming the JSON commands instead. `0.51.0` builds the reviewed rev-8 design: wordmark and project context, content capped at 96 columns, live rows naming the PARTICIPANT and their elapsed clock, spinners and verdict glyphs, breadcrumbs, the lab's subject/model/caps/keys line, and one Start with a dry-run/live toggle. `0.52.0` completes the reviewed screen set: the run outcome card (denominator first, the participant's closing words, then Open in Observer / Run again), the interrupted card with Reclaim — live-proven by killing a real run and stopping its orphaned sandbox — and All runs, the cross-lab peer where participants lead and one thought line follows the cursor. `0.53.0` gives it the reviewed palette rather than the terminal's theme: exact hex on a truecolor terminal, downsampled where not, and never colour alone | Cancelling a run from the surface; per-user persisted config (#470); export (#471); aggregate stats (#472); agent-authored labs (#473); a reusable persona panel (#474) |
+| Stakeholder terminal surface | `humanish tui` (`0.50.0`, redesigned to the reviewed spec in `0.51.0`): labs -> lab -> run, arrow-key navigation, and starting a dry or live run from the lab screen. The run is DETACHED and outlives the terminal — live-proven by killing the terminal 42s into a real run that then ran on for ~3.5 minutes and finished `pass` at $0.639751. The run screen leads with the participant and their recorded thinking, live-proven mid-flight against a real computer-use run. Ships as one bundled file loaded on demand; refuses a non-interactive stdin/stdout naming the JSON commands instead. `0.51.0` builds the reviewed rev-8 design: wordmark and project context, content capped at 96 columns, live rows naming the PARTICIPANT and their elapsed clock, spinners and verdict glyphs, breadcrumbs, the lab's subject/model/caps/keys line, and one Start with a dry-run/live toggle. `0.52.0` completes the reviewed screen set: the run outcome card (denominator first, the participant's closing words, then Open in Observer / Run again), the interrupted card with Reclaim — live-proven by killing a real run and stopping its orphaned sandbox — and All runs, the cross-lab peer where participants lead and one thought line follows the cursor. `0.53.0` gives it the reviewed palette rather than the terminal's theme: exact hex on a truecolor terminal, downsampled where not, and never colour alone. `0.54.0` closes the last two gaps: stopping a running run (armed, signals the process GROUP, and says plainly that sandboxes are separate), and pricing a run WHILE it runs — the running usage now travels with the trace into the mid-run flush | Cancelling a run from the surface; per-user persisted config (#470); export (#471); aggregate stats (#472); agent-authored labs (#473); a reusable persona panel (#474) |
 | Off-app comms | Vendor-neutral in-sandbox email/SMS catch, a minimal persona inbox surface, and digest-only `humanish.comms-thread.v1` evidence; wired into the computer-use and shared-world routes over both HTTP and SMTP; live-proven end to end on 2026-08-08 — a persona signed up for a public app, read the emailed link in its inbox, and reached the signed-in product (`docs/goals/email-gated-signup/receipts/signup-verify-live-2026-08-08.md`); the adopter-hosted / app-url ingress plane is wired on the CUA and concurrent external-public routes (#387/#380, 2026-08-11) | Real-provider delivery; a live adopter-hosted receipt |
 
 Capability proof and adopter replacement are different gates. A deterministic

package/docs/ramp/README.md

@@ -2,7 +2,7 @@
 
 Status: public-safe contributor and agent ramp.
 
-Package/source version in this tree: `0.53.0` (2026-08-20). The Observer is phone-usable as a stated requirement (observer/AGENTS.md); interactive primitives start from Base UI. The Observer renderer is the observer/ workspace artifact only; the legacy string-concat renderer was deleted at cutover (#426), and rollback is a version pin to 0.42.0. The containment boundary introduced in
+Package/source version in this tree: `0.55.0` (2026-08-20). The Observer is phone-usable as a stated requirement (observer/AGENTS.md); interactive primitives start from Base UI. The Observer renderer is the observer/ workspace artifact only; the legacy string-concat renderer was deleted at cutover (#426), and rollback is a version pin to 0.42.0. The containment boundary introduced in
 `0.15.1` remains in force: managed run and output paths bind to validated
 physical filesystem identities, and stored provider IDs are evidence, not
 cleanup authority. The bundled OSS meta-lab is dry-run only until

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "humanish",
-  "version": "0.53.0",
+  "version": "0.55.0",
   "description": "Open-source-safe CLI for persona simulation, observer review, and public-safe feedback drafts.",
   "author": "Daniel G Wilson <daniel@danielgwilson.com>",
   "keywords": [