humanish 0.15.2 → 0.16.0

package/docs/goals/current.md

@@ -1,6 +1,6 @@
 # Current Goals
 
-Status date: 2026-07-14 (rev 14)
+Status date: 2026-07-14 (rev 15)
 
 This page is the current public-safe operating goal for `humanish`. Keep it
 short enough to reread before a coding session and concrete enough that future
@@ -16,7 +16,39 @@ Humanish should be the open-source CLI that lets a maintainer ask:
 The answer should be observable, verifiable, public-safe, and easy to turn into
 actionable feedback.
 
-## Current Safety State (`0.15.1`)
+## Current Program Truth (source `0.16.0`)
+
+The package source and repository implementation in this tree agree on these
+points:
+
+The immutable 2026-06-10 proof-roadmap packet is paired with a
+[current implementation checkpoint](https://github.com/danielgwilson/humanish/blob/main/docs/goals/proof-roadmap/README.md).
+
+| 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 |
+| 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 | Version-pinned synthetic Observer sample shipped in the npm payload | A legible Observer hero from one verified real public-application run |
+| 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 |
+
+Capability proof and adopter replacement are different gates. A deterministic
+test or kept live receipt proves that a Humanish mechanism works. The depth-axis
+goal is met only when an adopter produces decision-equivalent evidence on a
+green branch that deletes its bespoke generic harness and retains at most a
+thin product-specific extension. No first-party deletion branch had met that
+bar as of this status date.
+
+Multi-origin shared-world has a ratified core-design direction, but the
+implementation gate is still closed. A real adopter must first show a concrete
+cross-origin need that the single-origin path or a downstream facade cannot
+serve cleanly; the implementation packet then requires maintainer review before
+build work starts. The current amendment is
+[`docs/goals/multi-origin-shared-world/README.md`](https://github.com/danielgwilson/humanish/blob/main/docs/goals/multi-origin-shared-world/README.md);
+the dated design packet remains unchanged.
+
+## Current Safety Boundary
 
 - Managed run, Observer, feedback, lab, actor-output, and source-archive paths
   bind to validated physical filesystem identities and fail closed on unsafe
@@ -141,9 +173,9 @@ Minimum acceptance:
   re-sequences the proof roadmap: a redaction redesign and an overridable
   public-target policy are prerequisites for any decision-grade depth evidence, so
   they land BEFORE the consumer-web-app / agent-skill depth phases. `done`
-- Device presets (0.6.1): viewport/device is a real dimension, with LITERAL values copied
+- Device presets (0.6.1): screen/device is a real dimension, with LITERAL values copied
   from the in-house sims (mobile 414×896 … wide 1920×1080; default `desktop` 1440×950) —
-  not guessed. `execution.desktop.device` picks the per-run viewport; the guessed 1280×800
+  not guessed. `execution.desktop.device` picks the per-run hosted screen; the guessed 1280×800
   is gone. Honest fidelity: on the E2B route only width/height render (real mobile *layout*)
   + the model is told its device, matching the sims' organic lanes; true touch/DPR/UA needs
   the CDP actor. Per-*persona* device (N×devices) rides fan-out. `done`
@@ -241,8 +273,9 @@ ONE shared world.
   host-side commit/dirty when the packed root is a git work tree; local-tree has no repo/publicRepo
   field. The N actor desktops on the concurrent route still drive the harness-minted getHost URL
   exactly as before; only the subject's provisioning + provenance source changed. The multi-origin
-  design (`docs/goals/multi-origin-shared-world/design.md`) remains a separate, still-held
-  downstream slice. `next patch`
+  design (`docs/goals/multi-origin-shared-world/design.md`) remains a separate,
+  ratified but implementation-gated downstream slice. It is not part of
+  `0.15.3`.
 
 Adopter-driven engine features (0.11.0; surfaced by real bespoke-sim migrations):
 
@@ -302,7 +335,7 @@ Evidence hygiene and readback polish (0.12.16):
   (`local_only`), while feedback draft/issue commands require `share_ready` and fail
   closed with structured reasons. `done`
 
-Attached CUA live Observer (next patch):
+Attached CUA live Observer (shipped):
 
 - Plain computer-use labs now honor the same attached `onObserverReady` lifecycle as shared-world
   labs: a live CUA run writes an in-progress bundle before actor sessions complete, loopback
@@ -356,7 +389,7 @@ Minimum acceptance:
 Make the maintainer `oss` lab report nested lane health back into the
 top-level Observer instead of relying on a human watching the desktops.
 
-The `0.15.1` safety state above governs this lane. The completed bullets below
+The current safety boundary above governs this lane. The completed bullets below
 record prior capability and evidence shape; they do not mean the live
 entrypoint is currently enabled.
 
@@ -414,15 +447,31 @@ Stop and correct course if:
 
 ## Best Next Work
 
-The next most useful engineering slice is repeated agent dogfood against real
-apps and tools, while preserving the public-safety boundary:
-
-- public/open-source fixture proof for publishable examples;
-- private maintainer dogfood through the repo-only public-safe packet, which is
-  intentionally not part of the npm payload, at
-  [`docs/goals/private-repo-agent-dogfood/goal.md`](https://github.com/danielgwilson/humanish/blob/main/docs/goals/private-repo-agent-dogfood/goal.md);
-- then richer provider-backed app-specific browser persona manifests.
-
-That sequence keeps the package honest: first prove a new maintainer or agent
-can start, then prove Humanish can observe real product behavior, then use the
-failures to improve the harness.
+After the public-truth reset, take one bounded public-proof side task: capture a
+verified, legible four-lane Observer from a commit-pinned public application.
+That run treats the application as a study subject, not a Humanish adopter, and
+must not imply endorsement. Stop and return to this order if the capture would
+require an Observer redesign or broader architecture work.
+
+The next engineering proof slice remains the first depth-axis deletion branch:
+run an existing first-party consumer web-app study through current Humanish,
+compare the decision outputs, and delete the adopter's bespoke generic harness
+while retaining only a thin product extension. The public-application capture
+does not satisfy this depth gate.
+
+After that checkpoint, re-evaluate this order rather than treating it as an
+automatic queue:
+
+1. centralize run identity, history, and provider ownership behind tested
+   `RunStore` and `ResourceLease` interfaces, and make generated first-run
+   source load-bearing or explicitly optional;
+2. complete the agent-skill deletion branch and use it to force the real
+   external extension budget;
+3. implement multi-origin shared-world only if its adopter gate opens and the
+   ratified packet passes maintainer review;
+4. complete the multi-app adopter deletion branch;
+5. establish the external stratified panel.
+
+The existing version-pinned README hero is a synthetic technical sample. It
+proves package/Observer rendering and public-safe asset delivery, not
+real-application evidence.

package/docs/ramp/README.md

@@ -2,9 +2,10 @@
 
 Status: public-safe contributor and agent ramp.
 
-Current safety state: `0.15.1` (2026-07-14). Managed run and output paths bind
-to validated physical filesystem identities. Stored provider IDs are evidence,
-not cleanup authority. The bundled OSS meta-lab is dry-run only until
+Package/source version in this tree: `0.16.0` (2026-08-01). 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
 repository-derived instructions have an isolated credential boundary.
 
 Use this page when you are starting cold on `humanish`. It is meant to be
@@ -69,13 +70,22 @@ Implemented:
 - first-class lab manifest resolution through `humanish/labs/*.yaml` and
   ignored `.humanish/labs/*.yaml` overlays — `humanish.lab.v2` compositions
   (`src/lab-config.ts`), one engine, no hardcoded lab kinds;
-- a pluggable actor registry with four registered actors
-  (`src/actor-registry.ts`); on the computer-use routes `actors[0].type` is a
-  real dispatch key;
+- a first-party actor registry with six registered descriptors
+  (`src/actor-registry.ts`); `actors[0].type` is a real dispatch key on the
+  computer-use, scripted-browser, and terminal-product routes;
 - a computer-use route and clone subject provider: `subject.source: app-url`
   drives a lab-owner loopback app in a hosted desktop, and `subject.source:
   clone` + `serve` clones, installs, and serves a real app in-sandbox from
   config before the actor drives it (`src/cua-actor-lab.ts`);
+- six declared subject sources: `this-repo` (dry-run-only), `clone`, `app-url`,
+  `local-app` (library-assisted, in-process, no desktop), `terminal-product`,
+  and `local-tree`; each route fails closed on unsupported combinations;
+- bounded per-lane-world fan-out (`actors[0].count`, `lanes[]`, or `roster[]`),
+  backed by deterministic and kept live proof;
+- sequential/concurrent single-origin shared-world execution: sequential has
+  deterministic proof, while concurrent has deterministic and kept live proof;
+- `subject.source: local-tree`, which packages one selected working tree with a
+  content pin before using the same provision-and-serve path as clone subjects;
 - containment checks for managed run storage, Observer and feedback reads,
   actor artifacts, lab discovery, Git metadata, and source archives;
 - an OSS meta-lab dry-run contract and a separate disposable public-repo OSS
@@ -85,16 +95,25 @@ Implemented:
 
 Still not good enough:
 
-- live `--app-url` browser proof now supports app-specific executable
-  `browser.steps` in `humanish/scenarios/*.yaml`, with headed provider-backed
-  public proof against two app/tool targets;
-- live PTY and Codex UI lanes need stronger completion health;
+The [current proof-roadmap checkpoint](https://github.com/danielgwilson/humanish/blob/main/docs/goals/proof-roadmap/README.md)
+supersedes implementation-status phrases in the immutable 2026-06-10 roadmap
+packet without changing its success standard.
+
+- capability receipts are not adopter replacement: no first-party deletion
+  branch has yet removed a bespoke generic harness while preserving
+  decision-equivalent proof;
+- the six actor descriptors are a closed first-party union, not a supported
+  out-of-tree actor-registration API;
+- run storage and provider-resource lifecycle logic still spans several routes
+  instead of one `RunStore` and `ResourceLease` boundary;
+- multi-origin shared-world is a ratified design direction, but remains
+  unimplemented and gated on a real adopter proving the need;
 - live OSS meta-lab execution remains disabled until repository-derived
   instructions have an isolated credential boundary; historical headed-lane
   evidence does not make the current entrypoint available;
-- Observer evidence has real screenshots/traces for browser app proof; richer
-  multi-step product journeys and broader multi-persona matrices remain the next
-  gap.
+- the README hero is a shipped, synthetic four-lane technical sample. It is not
+  proof of a real application study; a legible real-application capture remains open
+  product proof.
 
 ## First Commands
 

package/docs/release/open-source-readiness.md

@@ -81,7 +81,8 @@ from a fresh clone:
 - history scans have no private upstream system names, absolute maintainer paths,
   secret patterns, or generated runtime bundles;
 - reachable commit author and committer emails are GitHub noreply-style
-  addresses or explicitly approved public maintainer emails;
+  addresses (including GitHub's documented `ID+USERNAME` and username-only
+  forms) or explicitly approved public maintainer emails;
 - GitHub issues, PRs, labels, and project fields have been scanned or rewritten
   for public-safe language.
 
@@ -172,8 +173,8 @@ order is load-bearing (the OIDC binding is exact owner/repo + workflow path):
 A tag pushed before steps 1-2 fails auth (OIDC repository claim mismatch) or
 provenance validation (package.json repository.url vs actual workflow repo).
 
-Trusted Publishing should be configured for GitHub Actions before cutting the
-next tag:
+Trusted Publishing is configured for GitHub Actions. Verify these exact binding
+fields before each tag and re-point them after any repository rename:
 
 - provider: GitHub Actions
 - repository owner: `danielgwilson`
@@ -191,7 +192,7 @@ The workflow uses:
 - `npm publish --access public`;
 - no long-lived npm token secret.
 
-Future automated release flow after trusted publishing is configured:
+Automated release flow:
 
 ```bash
 pnpm release:check

package/docs/release/public-readiness-standard.md

@@ -10,6 +10,7 @@ durable context future contributors and agents need.
 
 - [GitHub Docs: Removing sensitive data from a repository](https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/removing-sensitive-data-from-a-repository)
 - [GitHub Docs: Setting your commit email address](https://docs.github.com/en/account-and-profile/how-tos/email-preferences/setting-your-commit-email-address)
+- [GitHub Docs: Email addresses reference](https://docs.github.com/en/account-and-profile/reference/email-addresses-reference)
 - [GitHub Docs: About secret scanning](https://docs.github.com/code-security/secret-scanning/about-secret-scanning)
 - [GitHub Docs: About push protection](https://docs.github.com/en/code-security/concepts/secret-security/about-push-protection)
 - [npm Docs: package.json files field](https://docs.npmjs.com/cli/v11/configuring-npm/package-json/#files)
@@ -103,7 +104,11 @@ Noreply commit emails are preferred for privacy and consistency.
 
 Allowed commit metadata:
 
-- GitHub noreply addresses.
+- GitHub's two documented personal-account noreply forms:
+  `ID+USERNAME@users.noreply.github.com` and
+  `USERNAME@users.noreply.github.com`. The username-only form remains valid for
+  accounts using GitHub's pre-July 18, 2017 privacy form.
+- `github-actions[bot]@users.noreply.github.com` for GitHub Actions commits.
 - `noreply@github.com` for GitHub-generated commits.
 - Explicitly approved public maintainer emails.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "humanish",
-  "version": "0.15.2",
+  "version": "0.16.0",
   "description": "Open-source-safe CLI for persona simulation, observer review, and public-safe feedback drafts.",
   "author": "Daniel G Wilson <daniel@danielgwilson.com>",
   "keywords": [