@junghanacs/entwurf 0.20.0 → 0.21.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (112) hide show
  1. package/AGENTS.md +85 -190
  2. package/BASELINE.md +6 -5
  3. package/CHANGELOG.md +220 -14
  4. package/CONTRIBUTING.md +1 -1
  5. package/DELIVERY.md +202 -60
  6. package/README.md +54 -26
  7. package/VERIFY.md +37 -7
  8. package/docs/acp-backend-rail.md +31 -15
  9. package/docs/external-mcp-host.md +58 -35
  10. package/docs/fresh-cut-policy.md +3 -2
  11. package/docs/setup-clean-host.md +110 -17
  12. package/mcp/entwurf-bridge/dist/mcp/entwurf-bridge/src/index.js +101 -109
  13. package/mcp/entwurf-bridge/dist/pi-extensions/lib/acp/backend-adapter.js +2 -2
  14. package/mcp/entwurf-bridge/dist/pi-extensions/lib/acp/tool-surface.js +6 -0
  15. package/mcp/entwurf-bridge/dist/pi-extensions/lib/codex-fresh-preflight.js +326 -0
  16. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-fact-provider.js +22 -2
  17. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-peers-render.js +3 -1
  18. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-v2-contract.js +6 -7
  19. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-v2-native-push.js +30 -17
  20. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-v2-production.js +7 -1
  21. package/mcp/entwurf-bridge/dist/pi-extensions/lib/entwurf-v2-surface.js +7 -4
  22. package/mcp/entwurf-bridge/dist/pi-extensions/lib/meta-mailbox-body.js +20 -11
  23. package/mcp/entwurf-bridge/dist/pi-extensions/lib/meta-sender-identity.js +227 -1
  24. package/mcp/entwurf-bridge/dist/pi-extensions/lib/mux-fresh-call.js +50 -15
  25. package/mcp/entwurf-bridge/dist/pi-extensions/lib/native-push/adapter.js +10 -7
  26. package/mcp/entwurf-bridge/dist/pi-extensions/lib/native-push/codex-ws-client.js +403 -0
  27. package/mcp/entwurf-bridge/dist/pi-extensions/lib/native-push/register.js +4 -4
  28. package/mcp/entwurf-bridge/dist/pi-extensions/meta-bridge-hook-codex.js +325 -0
  29. package/mcp/entwurf-bridge/dist/scripts/meta-bridge-fresh-cut.js +6 -1
  30. package/mcp/entwurf-bridge/src/index.ts +115 -111
  31. package/mcp/entwurf-bridge/tsconfig.build.json +1 -0
  32. package/package.json +14 -5
  33. package/pi-extensions/entwurf-control.ts +71 -19
  34. package/pi-extensions/lib/acp/acp-client.ts +3 -3
  35. package/pi-extensions/lib/acp/backend-adapter.ts +3 -3
  36. package/pi-extensions/lib/acp/backend.ts +3 -3
  37. package/pi-extensions/lib/acp/event-mapper.ts +4 -4
  38. package/pi-extensions/lib/acp/tool-surface.ts +6 -0
  39. package/pi-extensions/lib/codex-fresh-preflight.ts +363 -0
  40. package/pi-extensions/lib/compaction-send-guard.ts +80 -0
  41. package/pi-extensions/lib/entwurf-fact-provider.ts +29 -3
  42. package/pi-extensions/lib/entwurf-peers-render.ts +4 -1
  43. package/pi-extensions/lib/entwurf-v2-contract.ts +6 -7
  44. package/pi-extensions/lib/entwurf-v2-native-push.ts +35 -18
  45. package/pi-extensions/lib/entwurf-v2-production.ts +10 -3
  46. package/pi-extensions/lib/entwurf-v2-surface.ts +7 -4
  47. package/pi-extensions/lib/meta-mailbox-body.ts +22 -13
  48. package/pi-extensions/lib/meta-sender-identity.ts +305 -0
  49. package/pi-extensions/lib/mux-fresh-call.ts +64 -19
  50. package/pi-extensions/lib/native-push/adapter.ts +21 -24
  51. package/pi-extensions/lib/native-push/codex-ws-client.ts +506 -0
  52. package/pi-extensions/lib/native-push/register.ts +7 -9
  53. package/pi-extensions/meta-bridge-hook-codex.ts +371 -0
  54. package/run.sh +251 -35
  55. package/scripts/check-acp-usage-accounting.ts +9 -9
  56. package/scripts/check-agy-sender-identity.ts +1 -1
  57. package/scripts/check-codex-birth-hook.ts +264 -0
  58. package/scripts/check-codex-bridge-identity.ts +179 -0
  59. package/scripts/check-codex-native-push.ts +386 -0
  60. package/scripts/check-codex-sender-identity.ts +495 -0
  61. package/scripts/check-compaction-send-guard.ts +130 -0
  62. package/scripts/check-copilot-receive-arm.ts +4 -1
  63. package/scripts/check-entwurf-fact-provider.ts +38 -0
  64. package/scripts/check-entwurf-peers-surface.ts +13 -1
  65. package/scripts/check-entwurf-self-address.ts +15 -16
  66. package/scripts/check-entwurf-v2-contract.ts +4 -3
  67. package/scripts/check-entwurf-v2-decider.ts +7 -5
  68. package/scripts/check-entwurf-v2-native-push.ts +35 -7
  69. package/scripts/check-entwurf-v2-production.ts +203 -11
  70. package/scripts/check-entwurf-v2-runner.ts +1 -1
  71. package/scripts/check-entwurf-v2-surface.ts +1 -1
  72. package/scripts/check-gate-qualification.ts +7 -4
  73. package/scripts/check-harness-admission-parity.ts +0 -1
  74. package/scripts/check-install-surface.ts +23 -7
  75. package/scripts/check-mux-launch-tmux.ts +47 -2
  76. package/scripts/check-native-push-adapter.ts +20 -16
  77. package/scripts/check-native-push-register.ts +5 -1
  78. package/scripts/check-release-gate-outcomes.ts +47 -1
  79. package/scripts/check-setup-qualification.sh +3 -1
  80. package/scripts/codex-birth-doctor.sh +276 -0
  81. package/scripts/codex-birth-install.sh +414 -0
  82. package/scripts/codex-birth-uninstall.sh +170 -0
  83. package/scripts/codex-mcp-config.py +435 -0
  84. package/scripts/codex-statusline-config.py +434 -0
  85. package/scripts/codex_toml_io.py +532 -0
  86. package/scripts/lib/codex-fresh-live-protocol.ts +113 -0
  87. package/scripts/lib/codex-fresh-source-receipts.ts +399 -0
  88. package/scripts/lib/launch-receipt-windows.ts +46 -0
  89. package/scripts/lib/tmux-coordinate-row.ts +58 -0
  90. package/scripts/meta-bridge-fresh-cut.ts +6 -1
  91. package/scripts/mutants/codex-native.json +838 -0
  92. package/scripts/mutants/compaction-send-guard.json +103 -0
  93. package/scripts/mutants/entwurf-peers.json +19 -0
  94. package/scripts/mutants/mux-fresh-call.json +19 -8
  95. package/scripts/mutants/omp-fresh.json +6 -4
  96. package/scripts/mutants/pi-package-ownership.json +26 -0
  97. package/scripts/mutants/release-gate.json +13 -0
  98. package/scripts/mutants/v2-surface.json +53 -1
  99. package/scripts/pi_settings_io.py +3 -1
  100. package/scripts/raw-async-delivery/README.md +2 -1
  101. package/scripts/raw-codex-measure/README.md +114 -46
  102. package/scripts/register-pi-package.py +38 -30
  103. package/scripts/register-pi-provider.py +3 -2
  104. package/scripts/smoke-acp-raw-turn-live.ts +1 -1
  105. package/scripts/smoke-agy-native-push-live.ts +3 -1
  106. package/scripts/smoke-codex-birth.sh +347 -0
  107. package/scripts/smoke-codex-config-state.sh +511 -0
  108. package/scripts/smoke-codex-fresh-live.ts +1186 -0
  109. package/scripts/smoke-codex-native-push-live.ts +75 -0
  110. package/scripts/smoke-setup-verdict.sh +123 -10
  111. package/scripts/smoke-user-scope-citizen.sh +199 -71
  112. package/scripts/tsconfig.json +1 -0
package/README.md CHANGED
@@ -10,7 +10,7 @@ npm package: <https://www.npmjs.com/package/@junghanacs/entwurf>
10
10
 
11
11
  Legacy package: [`@junghanacs/pi-shell-acp`](https://www.npmjs.com/package/@junghanacs/pi-shell-acp). `entwurf` is its 0.12+ successor line: the same work renamed around the garden-citizen dispatch substrate rather than the pi adapter.
12
12
 
13
- > **Repository shape.** This repo is **entwurf-core (v2 dispatch) + native-harness bridges + a pi adapter + an ACP plugin**. Pi is one supported harness adapter — important because it supplies control sockets and hosts the ACP plugin today — but it is not the project subject. Claude Code and GitHub Copilot CLI are shipped as mailbox-backed self-fetch meta-sessions; Antigravity (`agy`) is shipped as a native-push citizen with automatic `PreInvocation` birth, ambient garden-id status, and a managed MCP/permission install surface. OMP (`omp`) is a self-fetch citizen on the same rail as Claude and Copilot, opened by `entwurf_fresh_call` and accepted under the step 9 visible-fresh contract on 2026-08-30 its first turn is a two-stage in-process bootstrap rather than an argv prompt, because the vendor connects its MCP tools in the background after the session starts. Codex has a launch-mode-specific verified delivery probe documented in [DELIVERY.md](./DELIVERY.md), but no managed native-citizen install lane yet. The ACP plugin ships two backends through one adapter rail: Claude (the reference) and Snowflake Cortex Code (landed in 0.13.0 under the measured contract in [docs/acp-backend-rail.md](./docs/acp-backend-rail.md#cortex-code-audit-d1d10)).
13
+ > **Repository shape.** This repo is **entwurf-core (v2 dispatch) + native-harness bridges + a pi adapter + an ACP plugin**. Pi is one adapter, not the project subject. Claude Code, GitHub Copilot CLI, and OMP (`omp`) are shipped self-fetch citizens; Antigravity (`agy`) is shipped native-push. Codex CLI is a **native-push citizen, supported in 0.21.0 on Linux** over the operator-owned app-server: vendor-trusted user-scope birth, strict request-scoped identity, loaded-thread probe, one-shot queue delivery, and visible fresh in one explicit operator-owned tmux home named `codex`. Unrestricted attached-TUI seat inference is not supported or claimed; that non-goal no longer blocks the explicit home topology. Codex remains native so it keeps its vendor tools, delegation, and work context; this is not another way to buy or expose GPT through ACP.
14
14
 
15
15
  ```text
16
16
  Claude Code / Copilot / Codex / agy / omp / pi
@@ -19,7 +19,7 @@ Claude Code / Copilot / Codex / agy / omp / pi
19
19
  → control-socket | meta-mailbox | native-push
20
20
  ```
21
21
 
22
- [`entwurf_v2`](#entwurf_v2--canonical-dispatch-verb) is the canonical dispatch surface over *existing* garden citizens — live control-socket send, meta-mailbox enqueue, and native-push into a live Antigravity conversation. It starts no process on any rail: the hidden background resume that used to answer a dormant target was withdrawn under the visible-first rule, so a dormant citizen rejects honestly here and is reopened by a separate lifecycle verb, `entwurf_resume_call`. The meta-record is the sole address authority (#50 C4): a record-less control socket is refused as a `record-less-socket` diagnostic, never dispatched. The v1 entwurf verbs are gone. Fresh siblings and resumes are separate verbs — `entwurf_fresh_call` opens a NEW sibling in the operator's tmux the caller's own session, or one existing named session on the same server — and learns its garden id from the callback it makes, while `entwurf_resume_call` reopens a DORMANT pi citizen under its own garden id in a visible window without running a turn; the non-Claude ACP lane landed earlier Snowflake Cortex Code became the second backend in 0.13.0.
22
+ [`entwurf_v2`](#entwurf_v2--canonical-dispatch-verb) is the canonical dispatch surface over *existing* garden citizens — live control-socket send, meta-mailbox enqueue, and native-push into a live Antigravity conversation or loaded Codex thread. It starts no process on any rail. `entwurf_fresh_call` is separate lifecycle. With Codex placement omitted, it resolves the exact existing tmux session named `codex`; that operator-owned home holds the app-server and supported Codex TUIs. Missing home or app-server rejects before launch, and Entwurf never creates or supervises either. An explicit `placement.tmuxSession` remains an expert override. Exact 0.153.4 source confirms that arbitrary attached-TUI request→seat inference is unavailable, so Entwurf makes no such claim. Codex has no resume surface.
23
23
 
24
24
  **Garden id is deliberate vocabulary.** It is not a decorative synonym for session id, worker, delegate, or subagent. The unfamiliar word is a guard: each harness keeps its own identity and transcript, while `entwurf` supplies a narrow addressable surface between siblings.
25
25
 
@@ -60,10 +60,9 @@ native Antigravity / agy
60
60
  ```
61
61
 
62
62
  Claude's `install-meta-bridge`, Copilot's four `install-copilot-*` surfaces, agy's
63
- `install-agy-{bridge,statusline,hooks}` and OMP's four `install-omp-{bridge,mcp,config,receive}`
64
- units are distinct because their lifecycle and delivery transports are
65
- genuinely different. Codex remains verified probe evidence, not a shipped managed native-citizen
66
- lane; see [DELIVERY.md](./DELIVERY.md).
63
+ `install-agy-{bridge,statusline,hooks}`, OMP's four `install-omp-{bridge,mcp,config,receive}`,
64
+ and Codex's birth, MCP and statusline units are distinct because their
65
+ lifecycle and delivery transports are genuinely different. See [DELIVERY.md](./DELIVERY.md).
67
66
 
68
67
  > **Direction.** Inverse of [`pi-acp`](https://github.com/svkozak/pi-acp). `pi-acp` lets external ACP clients talk *to* pi; `entwurf` lets garden citizens talk across harness boundaries — with pi as one adapter, not the center.
69
68
 
@@ -72,7 +71,7 @@ lane; see [DELIVERY.md](./DELIVERY.md).
72
71
  A few words that look unusual for a coding tool.
73
72
 
74
73
  - **Entwurf** (기투, projection-of-self) — sibling sessions with their own runtime boundary. Not "delegate," not "worker," not "sub-agent." Opening a visible sibling (`entwurf_fresh_call`), live peer messaging (`entwurf_v2`) and reopening a dormant one (`entwurf_resume_call`) are first-class; the hidden background resume that preceded the last of those was withdrawn under the visible-first rule.
75
- - **Garden / garden id** — the garden is the shared address space where independent harness sessions become citizens without losing their own runtime or transcript. A garden id is the stable address of one such citizen (for pi, a garden-native session id like `YYYYMMDDTHHMMSS-<6hex>`; for native harnesses, a meta-session id minted from an authoritative lifecycle hook — Claude `SessionStart`, Copilot's first-prompt birth hook, agy `PreInvocation`, and for OMP an in-process extension bound to both session edges that mints only the visible `mode === "tui"` host). It is not a worker name and not proof that pi owns the session. The same-looking id may name a live control socket, a dormant pi record, a mailbox-backed native session, or a native-push conversation, so callers discover facts with `entwurf_peers` and deliver with `entwurf_v2` instead of choosing a transport by hand.
74
+ - **Garden / garden id** — the garden is the shared address space where independent harness sessions become citizens without losing their own runtime or transcript. A garden id is the stable address of one such citizen (for pi, a garden-native session id like `YYYYMMDDTHHMMSS-<6hex>`; for native harnesses, a meta-session id minted from an authoritative lifecycle hook — Claude `SessionStart`, Codex `SessionStart`, Copilot's first-prompt birth hook, agy `PreInvocation`, and for OMP an in-process extension bound to both session edges that mints only the visible `mode === "tui"` host). It is not a worker name and not proof that pi owns the session. The same-looking id may name a live control socket, a dormant pi record, a mailbox-backed native session, or a native-push conversation/thread, so routing always reads the record and live rail facts first.
76
75
  - **Engraving** — optional short operator text delivered through each backend's native identity carrier. Not a giant hidden prompt, not a tool catalog.
77
76
  - **MCP** — in this repo, MCP is just the transport by which ACP-backed sessions receive pi capabilities that native pi exposes directly as extensions. It is not a general MCP platform. Explicit `entwurfProvider.mcpServers` only; no ambient `~/.mcp.json` scanning, no automatic retrieval. The same `entwurf-bridge` entry can also be wired into another host's MCP catalog (Claude Code, Copilot, Codex, Antigravity, OMP, …) when the operator chooses. `entwurf_self` returns an authoritative pi-session or trusted meta-session identity envelope; `entwurf_v2` requires an authoritative sender by default (#50 C4) — a plain external MCP host with no identity lane is refused unless the operator explicitly wires the documented anonymous hatch, and even then it is never replyable.
78
77
  - **Session persistence** — re-attaches pi to the same remote ACP session. Does not hydrate backend transcripts into pi history.
@@ -81,21 +80,28 @@ A few words that look unusual for a coding tool.
81
80
 
82
81
  **Platform evidence, in one breath.** The Entwurf-only install surface on
83
82
  macOS is CERTIFIED (CI) (`macos-install-surface`). On macOS every
84
- garden-native harness rail (pi, Claude, Copilot, OMP, agy), marker join,
85
- ACP turn, and mux is NOT CERTIFIED pending physical host; on Linux those
86
- same rails are the certified axis. native Windows is UNSUPPORTED.
83
+ garden-native harness rail, marker/request join, ACP turn, and mux is NOT CERTIFIED —
84
+ pending physical host. On Linux the released harness rails are the certified axis;
85
+ Codex joined them in 0.21.0. Its explicit-home first admission was measured on 2026-09-12 — a real Pi
86
+ outside `codex` opened Codex and outbound Pi inside it, with exact callbacks and delivery both ways —
87
+ and the release acceptance that followed closed `check:full`, the LIVE release gate at MUST 24/0/0,
88
+ and a 475/475 qualification.
89
+ Unrestricted attached-TUI placement
90
+ is unsupported rather than a release blocker.
91
+ Native Windows is UNSUPPORTED.
87
92
  CERTIFIED (CI) is weaker than a physical-host doctor green.
88
93
  The npm package has no `os` restriction; that is installability, not a
89
94
  support claim.
90
95
 
91
- **macOS in this release.** This release adds macOS support for Entwurf's
96
+ **macOS landed in 0.20.0.** That release added macOS support for Entwurf's
92
97
  install surface: the npm package installs, `entwurf --help` and `entwurf
93
98
  check-bridge` boot, and `entwurf setup` composes the integrations it finds.
94
99
  The four installers that previously rejected Darwin by platform name no longer
95
- do. We validated that surface only in macOS CI because the developers did not
100
+ do. That surface was validated only in macOS CI because the developers did not
96
101
  have a physical Mac. So macOS native-harness rails—marker join, an ACP turn,
97
- and mux included—are not yet certified; setup writes their wiring but reports
98
- them non-green. Linux's certified rail status is unchanged.
102
+ and mux included—are **NOT CERTIFIED pending physical host**; setup writes
103
+ their wiring but reports them non-green. Linux's certified rail status is
104
+ unchanged.
99
105
 
100
106
  On Darwin, `entwurf setup` that detects a harness is intentionally
101
107
  non-green: the wiring WAS written and nothing failed to install, but the
@@ -242,15 +248,18 @@ SKIP while the detected harnesses are composed.
242
248
  ### Native harness repair and doctors
243
249
 
244
250
  A plain MCP registration exposes the bridge tools; a **garden-native** session also
245
- needs entwurf's lifecycle hook and identity marker. `setup` already composes all of that
246
- for every harness it detects you do not paste this list to install. This is the repair
247
- surface: each unit has its own installer, its own doctor with a named refusal, and its own
248
- inverse, so a single broken unit can be redone without touching the rest.
251
+ needs entwurf's lifecycle and identity unit. `setup` composes every user-scope unit
252
+ for each detected harness, including Codex's birth hook its paths are the operator's
253
+ own, so nothing here asks for root. What setup cannot do is answer the vendor's one-time
254
+ hook-trust prompt: only the operator can, in their own visible Codex, so a first setup on
255
+ a Codex host writes every byte and stays honestly non-green until they have. This is the
256
+ repair surface: each unit has its own installer, doctor, and inverse.
249
257
 
250
258
  - **Claude Code** (Linux CERTIFIED; macOS NOT CERTIFIED — pending physical host) — `install-meta-bridge`, `doctor-meta-bridge`.
251
259
  - **Antigravity / agy** (Linux CERTIFIED; macOS NOT CERTIFIED — pending physical host) — `install-agy-bridge`, `install-agy-statusline`, `install-agy-hooks`, each with a matching `doctor-agy-*`.
252
260
  - **GitHub Copilot CLI** (Linux CERTIFIED; macOS NOT CERTIFIED — pending physical host) — four independent units, four independent failure modes: `install-copilot-bridge` (birth: garden id + who-sent, on the first prompt), `install-copilot-mcp` (the entwurf tool hand, where `entwurf_inbox_read` lives), `install-copilot-receive` (the receiver extension: doorbell + receiver marker), `install-copilot-statusline` (optional for a manual citizen, required for supported fresh) — each with a matching `doctor-copilot-*` and `uninstall-copilot-*`.
253
261
  - **OMP (`omp`)** (Linux CERTIFIED; macOS NOT CERTIFIED — pending physical host) — four units, in-process extensions rather than launchers: `install-omp-bridge` (birth: the `mode === "tui"` visible host, its garden id on the status line, and who-sent), `install-omp-mcp` (the omp-native `entwurf-bridge` entry), `install-omp-config` (the one operator setting `tools: xdev: false`, without which the vendor mounts MCP tools as `xd://` devices the model cannot call), `install-omp-receive` (the receiver extension: mailbox watch + announce-only doorbell) — each with a matching `uninstall-omp-*`, and a `doctor-omp-*` for all but the setting, whose runtime axis `doctor-omp-mcp` owns. The setting writer owns exactly the lines it adds and refuses an explicit operator `tools: xdev: true` by name rather than overwriting it.
262
+ - **OpenAI Codex CLI** (supported in 0.21.0 on Linux; macOS NOT CERTIFIED — pending physical host) — `install-codex-birth` publishes the `SessionStart` declaration into `$CODEX_HOME/hooks.json` with its launcher closure under `$XDG_DATA_HOME/entwurf/codex-birth`, all operator-owned; the vendor trust receipt for that declaration is the operator's single answer, and `doctor-codex-birth` reports it as its own axis (present, or red with the exact instruction — never computed, never written); `install-codex-mcp` owns `[mcp_servers.entwurf-bridge]`, including the `env_vars` boundary for `CODEX_HOME`, Entwurf garden/control roots, and the app-server's `TMUX`/`TMUX_PANE`; `install-codex-statusline` owns `thread-title`. Each has a matching doctor and inverse. The operator owns one existing tmux session named `codex`, starts the app-server there, and seats supported Codex TUIs there. Omitted Codex fresh placement selects that home; Entwurf never creates or supervises the session/app-server and never discovers arbitrary attached-TUI seats. An explicit placement is an expert override. Exact 0.153.4 source's missing per-client carrier bounds the unsupported topology; it does not block this explicit home deployment.
254
263
 
255
264
  Run them as `entwurf <command>`. Which unit a doctor's refusal names, and the clean-host
256
265
  walk-through for each harness, live in [docs/setup-clean-host.md](./docs/setup-clean-host.md).
@@ -335,9 +344,9 @@ host evidence boundaries are [VERIFY.md](./VERIFY.md) and [BASELINE.md](./BASELI
335
344
 
336
345
  The curated model registry exposes unprefixed Claude ids — `claude-opus-5`, `claude-sonnet-5`,
337
346
  `claude-fable-5-1` — plus the `cortex-` rows below.
338
- Codex is not an ACP backend or a shipped managed citizen lane: it has verified
339
- native-delivery probe evidence only. No managed Codex support is claimed; lifecycle,
340
- identity, installation, and doctors remain prerequisites for any such lane.
347
+ Codex is not an ACP backend. Its native-citizen lane, supported in 0.21.0, is independent so the
348
+ session retains Codex's native tools, delegation, and work context: trust-gated hook birth, strict
349
+ request `_meta` identity, app-server probe, one-shot native-push, and visible fresh.
341
350
 
342
351
  **Snowflake Cortex Code is the second ACP backend** (contract and audit:
343
352
  [docs/acp-backend-rail.md](./docs/acp-backend-rail.md#cortex-code-audit-d1d10)). Curated ids are
@@ -443,8 +452,10 @@ The Claude ACP backend keeps its native model / API / tools; entwurf shapes only
443
452
 
444
453
  **Claude** uses `_meta.systemPrompt` for the engraving carrier (kept short and pure — billing-safe; rich operator context rides the first user message instead, see [Context carriers](#context-carriers)) and `CLAUDE_CONFIG_DIR` for a whitelist overlay so auth/runtime entries stay available while operator memory, hooks, agents, history, local settings, and project memory remain hidden. The overlay writes an explicit empty `hooks: {}` because Claude SDK organic compaction needs the configured-empty shape; no operator hook definitions are inherited. It also pins `permissions.defaultMode: "bypassPermissions"` so an unattended ACP turn cannot suspend on an interactive permission prompt; explicit `tools` / `disallowedTools` still constrain the callable surface and backend authentication remains the operator's. The four-tool baseline is `Read`, `Bash`, `Edit`, and `Write`; `permissionAllow` carries their allow declarations, and `Skill` is added automatically when `skillPlugins` is non-empty. Operator context cap override: `ENTWURF_ACP_CLAUDE_CONTEXT=<int>`.
445
454
 
446
- Codex is not an ACP backend here. Its native delivery probe remains separate from
447
- the governed ACP adapter rail and does not yet constitute a managed garden citizen.
455
+ Codex is not an ACP backend here. Its native-push lane, supported in 0.21.0, uses `SessionStart`
456
+ for birth, strict request `_meta` for who-sent, and a loaded app-server thread as the live route.
457
+ Its supported visible topology is one operator-owned existing tmux home named `codex`; omitted
458
+ Codex fresh placement selects it, while unrestricted attached-TUI seat inference is not claimed.
448
459
 
449
460
  Antigravity is also not an ACP backend. It is a native-push citizen: `PreInvocation` supplies birth/sender identity, `entwurf_v2` probes and direct-injects replies into the live conversation, and no mailbox/receiver marker is involved.
450
461
 
@@ -460,6 +471,9 @@ pnpm run check:full # full deterministic floor (adds the her
460
471
  ./run.sh smoke-agy-statusline-state # agy ambient garden-id install surface
461
472
  ./run.sh smoke-agy-hooks-state # agy PreInvocation birth hook
462
473
  ./run.sh check-agy-sender-identity # record-backed pid/start-key sender identity
474
+ ./run.sh check-codex-native-push # deterministic Codex UDS probe + no-replay queue contract
475
+ ./run.sh smoke-codex-config-state # Codex MCP/status-line install, doctor, and inverse
476
+ ./run.sh check-codex-birth-hook # sandboxed payload + declaration: one V3 per top-level thread, the envelope's own axes, no markers
463
477
 
464
478
  # source-maintainer only — qualification snapshots the git work surface, and both
465
479
  # commands are source-contract gates rather than installed operator checks:
@@ -470,6 +484,20 @@ pnpm run check:full # full deterministic floor (adds the her
470
484
  # agy LIVE acceptance — requires an already-running conversation:
471
485
  LIVE=1 AGY_CONVERSATION_ID=<id> ./run.sh smoke-agy-native-push-live
472
486
 
487
+ # Codex loaded-thread probe (pre-amendment evidence; not first-release acceptance):
488
+ LIVE=1 CODEX_LIVE_THREAD_ID=<threadId> ./run.sh smoke-codex-native-push-live
489
+
490
+ # Codex explicit-home acceptance. Run from a DIFFERENT tmux session; the operator-owned
491
+ # app-server and supported Codex TUIs sit in the existing exact `codex` session. Record
492
+ # initial Pi, app-server, fresh Codex, and outbound Pi coordinates separately:
493
+ LIVE=1 ENTWURF_CODEX_APP_SERVER_PID=<existing-app-server-pid> \
494
+ ENTWURF_CODEX_FRESH_MODEL=<codex-model> \
495
+ ENTWURF_CODEX_FRESH_PI_MODEL=<pi-model> \
496
+ ./run.sh smoke-codex-fresh-live
497
+ # This accepts only a real visible Pi → visible Codex → visible Pi run: initial Pi outside
498
+ # the Codex home, then Codex and its outbound Pi inside it. A fixture or self-fetch first leg
499
+ # may collect receipts but does not satisfy the topology. No arbitrary attached-TUI parity is claimed.
500
+
473
501
  # ACP plugin LIVE acceptance — need the operator's local Claude auth/credit:
474
502
  LIVE=1 ./run.sh smoke-acp-socket-citizen-live # turn-free socket citizenship (S1)
475
503
  LIVE=1 ./run.sh smoke-acp-raw-turn-live # pinned ACP pipe + raw 1 turn (S2a)
@@ -544,7 +572,7 @@ A two-pane recording covers the pre-0.12 v1 surface end-to-end — sibling resum
544
572
 
545
573
  Live peer messaging carries a sender envelope `{ sessionId, agentId, cwd, timestamp, origin?, replyable? }`; `entwurf_self` returns that authoritative envelope for the current pi session or trusted meta-session. Plain external MCP hosts are non-replyable. A garden-native meta-session carries a trusted `meta-session` envelope, but **`replyable` is a fact its own rail decides, not a consequence of being trusted** — a self-fetch citizen needs a live armed receiver, a native-push citizen needs an alive adapter probe, and a pi session needs its control socket. `entwurf_self` also reports which rail a meta-session reply would ride, because a native-push citizen has no mailbox to name. `wants_reply` is an etiquette marker rendered as a `(wants reply)` badge — not a transport contract, no wait, no polling. **v2 never gates on it:** a `wants_reply` from an external/non-replyable caller is passed through and surfaced honestly beside that sender's `replyable: false`, not rejected — the decider routes on target + intent, never on sender replyability. (The retired v1 `entwurf_send` did reject it; that behaviour went with the verb.)
546
574
 
547
- In ACP-backed and external native-harness sessions, `entwurf-bridge` exposes seven tools: `entwurf_v2`, `entwurf_peers`, `entwurf_fresh_call`, `entwurf_resume_call`, `entwurf_self`, `entwurf_inbox_read`, and the explicit/manual `entwurf_register_native` fallback. Native pi exposes the shared capability directly through the extension surface (`entwurf_v2`, `entwurf_peers`, `entwurf_fresh_call`, `entwurf_resume_call` tools; the socket-scan `/entwurf-sessions` command is gone — #50 C4). **For garden-id delivery/reply use `entwurf_v2`** — the canonical surface that classifies the target and routes to live-pi / self-fetch meta-mailbox (Claude Code, Copilot, OMP) / Antigravity-native-push, and refuses a dormant target instead of waking it. **To open a sibling that does not exist yet use `entwurf_fresh_call`** — it launches one fixed backend (`pi`, `claude-code`, `copilot`, or `omp`) as a visible window on the operator's own tmux server, requires an explicit model, accepts one optional literal absolute `cwd` (omit it or pass `""` to use the caller's cwd) and one optional `placement.tmuxSession` naming an EXISTING session on that server to open the sibling in instead of the caller's own (the operator's project seat — nothing is ever created: an absent session is `tmux-session-missing` and a name outside the addressed grammar `[A-Za-z0-9][A-Za-z0-9_-]*` is `tmux-session-name-invalid`, with no window and no session left behind. The two inputs never infer each other, and the positive form of that matters more than the prohibition: **with `cwd` omitted, a seated sibling starts in THIS agent's own working directory**, measured — it does not inherit the target session's path or its active pane, so naming the `org` seat does not put the sibling in the `org` project. Pass `cwd` when you want a different directory. entwurf never resolves that path: it hands the literal string to tmux, so with a SYMLINKED directory the receipt echoes what you asked for while the record, the transcript path, the callback envelope and `pane_current_path` all show the path the OS resolved it to — two spellings of one place, and the resolved one is what `entwurf_resume_call` later reopens), passes the model and selected directory through the runtime's visible launch path, and hands it a first task; a `copilot` launch goes through entwurf's own managed invocation and is refused before any window opens if this host lacks the Copilot birth, MCP, receiver or visible-footer units, and an `omp` launch carries its task in the `--entwurf-bootstrap` payload instead of an argv prompt because that vendor connects its MCP tools after the session has already started. The sibling's first action is a nonce callback whose sender envelope carries its garden id. The launch receipt records the requested model/cwd/seat plus tmux coordinates — including the RESOLVED target session id, which is where the window actually is — and never claims that the runtime accepted them or completed delivery. (The v1 verbs `entwurf` / `entwurf_resume` / `entwurf_send` are gone.) Garden-native operator commands require `--entwurf-control`. There is no spawn target allowlist — the target registry is gone (#50 C3): `entwurf_v2` and `entwurf_resume_call` address an existing record-backed citizen, while `entwurf_fresh_call` takes its explicit backend/model/task and optional cwd directly rather than resolving a model tuple from a registry file. **To reopen a DORMANT pi citizen use `entwurf_resume_call {target}`** — the record supplies transcript, model, provider and cwd, so it takes no prompt, no task and no model override; it runs no turn, returns a LAUNCH receipt and a separate OBSERVATION receipt (only the second says the citizen is back), refuses a non-pi target as `target-not-pi`, and on an unobserved socket leaves the visible window open and releases its lock rather than retrying.
575
+ In ACP-backed and external native-harness sessions, `entwurf-bridge` exposes seven tools: `entwurf_v2`, `entwurf_peers`, `entwurf_fresh_call`, `entwurf_resume_call`, `entwurf_self`, `entwurf_inbox_read`, and the explicit/manual `entwurf_register_native` fallback. Native pi exposes the shared capability directly through the extension surface (`entwurf_v2`, `entwurf_peers`, `entwurf_fresh_call`, `entwurf_resume_call` tools; the socket-scan `/entwurf-sessions` command is gone — #50 C4). **For garden-id delivery/reply use `entwurf_v2`** — the canonical surface that classifies the target and routes to live-pi / self-fetch meta-mailbox (Claude Code, Copilot, OMP) / native-push (Antigravity, Codex), and refuses a dormant target instead of waking it. **To open a sibling that does not exist yet use `entwurf_fresh_call`** — it launches one of five fixed backends (`pi`, `claude-code`, `copilot`, `omp`, `codex`) as a visible window on the operator's own tmux server, requires an explicit model, accepts one optional literal absolute `cwd` (omit it or pass `""` to use the caller's cwd) and one optional `placement.tmuxSession` naming an EXISTING session on that server to open the sibling in instead of the caller's own (the operator's project seat — nothing is ever created: an absent session is `tmux-session-missing` and a name outside the addressed grammar `[A-Za-z0-9][A-Za-z0-9_-]*` is `tmux-session-name-invalid`, with no window and no session left behind. The two inputs never infer each other, and the positive form of that matters more than the prohibition: **with `cwd` omitted, a seated sibling starts in THIS agent's own working directory**, measured — it does not inherit the target session's path or its active pane, so naming the `org` seat does not put the sibling in the `org` project. Pass `cwd` when you want a different directory. entwurf never resolves that path: it hands the literal string to tmux, so with a SYMLINKED directory the receipt echoes what you asked for while the record, the transcript path, the callback envelope and `pane_current_path` all show the path the OS resolved it to — two spellings of one place, and the resolved one is what `entwurf_resume_call` later reopens), passes the model and selected directory through the runtime's visible launch path, and hands it a first task; a `copilot` launch goes through entwurf's own managed invocation and is refused before any window opens if this host lacks the Copilot birth, MCP, receiver or visible-footer units, and an `omp` launch carries its task in the `--entwurf-bootstrap` payload instead of an argv prompt because that vendor connects its MCP tools after the session has already started. Omitted placement is now a per-backend policy rather than one rule: `pi`, `claude-code`, `copilot` and `omp` open in the CALLER's own seat, while `codex` resolves the exact existing operator-owned `codex` session (a missing home rejects; nothing is created). 0.21.0 is the first release to carry such a backend-selected default — every backend could already be sent to an operator-owned existing seat with an explicit `placement.tmuxSession`, and named `claude`/`pi` seats remain available that way, but they are not omitted-placement defaults. Read tmux here as the VISIBLE LAUNCH SEAT, never as an address: the garden id is the address, and it arrives in the callback envelope. Sibling symmetry means each backend keeps its own transcript, auth, native tools and visibility — not that the transports are made identical. The sibling's first action is a nonce callback whose sender envelope carries its garden id. The launch receipt records the requested model/cwd/seat plus tmux coordinates — including the RESOLVED target session id, which is where the window actually is — and never claims that the runtime accepted them or completed delivery. (The v1 verbs `entwurf` / `entwurf_resume` / `entwurf_send` are gone.) Garden-native operator commands require `--entwurf-control`. There is no spawn target allowlist — the target registry is gone (#50 C3): `entwurf_v2` and `entwurf_resume_call` address an existing record-backed citizen, while `entwurf_fresh_call` takes its explicit backend/model/task and optional cwd directly rather than resolving a model tuple from a registry file. **To reopen a DORMANT pi citizen use `entwurf_resume_call {target}`** — the record supplies transcript, model, provider and cwd, so it takes no prompt, no task and no model override; it runs no turn, returns a LAUNCH receipt and a separate OBSERVATION receipt (only the second says the citizen is back), refuses a non-pi target as `target-not-pi`, and on an unobserved socket leaves the visible window open and releases its lock rather than retrying.
548
576
 
549
577
  ### `entwurf_v2` — canonical dispatch verb
550
578
 
@@ -561,9 +589,9 @@ In ACP-backed and external native-harness sessions, `entwurf-bridge` exposes sev
561
589
  | dead / indeterminate native-push conversation | fire-and-forget | **reject** (`native-push-target-dead` / `native-push-probe-indeterminate`) |
562
590
  | record-less control socket (no meta-record) | any | **reject** (`record-less-socket` — pre-probe; diagnostic state, #50 C4) |
563
591
 
564
- **`entwurf_v2` is the canonical surface for garden-id delivery.** When you have a garden id and want to reach whoever it names — message, reply, or hand-off — `entwurf_v2` is the one surface that reads whether the target is live pi, dormant pi, mailbox-backed Claude Code, Copilot or OMP, or native-push Antigravity and routes correctly; *when unsure which transport, use `entwurf_v2`*. This prevents callers from guessing a rail from the shape of an id.
592
+ **`entwurf_v2` is the canonical surface for garden-id delivery.** When you have a garden id and want to reach whoever it names — message, reply, or hand-off — `entwurf_v2` is the one surface that reads whether the target is live pi, dormant pi, mailbox-backed Claude Code, Copilot or OMP, or native-push Antigravity or Codex, and routes correctly; *when unsure which transport, use `entwurf_v2`*. This prevents callers from guessing a rail from the shape of an id.
565
593
 
566
- What v2 provides is a **deterministic dispatch substrate** that moves the "which transport?" decision out of the fallible caller/model and into the decider, with transport-appropriate locking and an honest reject (no `✓ delivered`, no `.msg` garbage) when a target cannot receive. It still does **not** mint siblings, and it does not relaunch one either: every row above either reaches a citizen that is already running or refuses. Reopening a dormant pi citizen is `entwurf_resume_call`, a separate lifecycle verb that never routes through this decider. Fresh creation is the separate `entwurf_fresh_call` verb. It opens one fixed Pi, Claude Code, Copilot or OMP runtime visibly on the caller's tmux server its own session by default, or one EXISTING named session via the optional `placement.tmuxSession` with a required explicit model passed in that runtime's measured CLI dialect and one optional literal absolute `cwd`; omitted or `""` means the caller's cwd. Copilot opens through `entwurf copilot`, never the bare vendor; OMP is the opposite — the bare `omp` runtime with no positional prompt at all, because that vendor connects its MCP tools after the session has started, so the task rides a two-stage `--entwurf-bootstrap` payload the installed birth extension releases once the callback tool has actually answered. It returns only a synchronous launch receipt and lets the sibling report its new address asynchronously through the sender envelope of a nonce callback. Use this cwd input for a new cross-repository sibling; do not resume a dormant citizen as a cwd substitute. The meta-mailbox row requires an **active** self-fetch receiver; native-push requires a record-backed, probe-alive native conversation and never borrows mailbox state. The [mux launch lane](./docs/mux-launch-rail.md) owns placement, fixed-runtime launch, and the two narrow compositions above it (fresh-call and resume-call placement); delivery does not import launch, and mux is not a delivery transport.
594
+ What v2 provides is a **deterministic dispatch substrate** that moves the "which transport?" decision out of the fallible caller/model and into the decider, with transport-appropriate locking and an honest reject (no `✓ delivered`, no `.msg` garbage) when a target cannot receive. It still does **not** mint siblings, and it does not relaunch one either: every row above either reaches a citizen that is already running or refuses. Reopening a dormant pi citizen is `entwurf_resume_call`, a separate lifecycle verb that never routes through this decider. Fresh creation is the separate `entwurf_fresh_call` verb. It opens one of five fixed runtimes — Pi, Claude Code, Copilot, OMP or Codex visibly on the caller's tmux server. Omitting placement is a per-backend policy: the first four open in the caller's own session, while Codex resolves the exact EXISTING operator-owned `codex` home and rejects if it is absent. An explicit `placement.tmuxSession` sends ANY of the five to an existing named session instead (see the `entwurf_fresh_call` paragraph above for the owning detail). Each launch carries a required explicit model passed in that runtime's measured CLI dialect and one optional literal absolute `cwd`; omitted or `""` means the caller's cwd. Copilot opens through `entwurf copilot`, never the bare vendor; OMP is the opposite — the bare `omp` runtime with no positional prompt at all, because that vendor connects its MCP tools after the session has started, so the task rides a two-stage `--entwurf-bootstrap` payload the installed birth extension releases once the callback tool has actually answered. It returns only a synchronous launch receipt and lets the sibling report its new address asynchronously through the sender envelope of a nonce callback. Use this cwd input for a new cross-repository sibling; do not resume a dormant citizen as a cwd substitute. The meta-mailbox row requires an **active** self-fetch receiver; native-push requires a record-backed, probe-alive native conversation and never borrows mailbox state. The [mux launch lane](./docs/mux-launch-rail.md) owns placement, fixed-runtime launch, and the two narrow compositions above it (fresh-call and resume-call placement); delivery does not import launch, and mux is not a delivery transport.
567
595
 
568
596
  A live pi target is *reached* over its control socket, but the socket is dispatch-internal transport, never identity (#50 C4). A control socket that no meta-record claims — a pre-record-era resident, an unreadable store, or a stale/planted file — is refused for **every** intent as `record-less-socket`, and the reject names the fix (restart the resident so `session_start` births its record, or quiesce and run the fresh-cut). `entwurf_peers` reports the same state as an aggregated `record-less-socket` diagnostic rather than a peer row.
569
597
 
package/VERIFY.md CHANGED
@@ -7,8 +7,14 @@ and the manual judgements a gate cannot make.
7
7
  > **Current surface.** `entwurf-bridge` exposes `entwurf_v2`, `entwurf_peers`,
8
8
  > `entwurf_fresh_call`, `entwurf_resume_call`, `entwurf_self`, `entwurf_inbox_read`, and `entwurf_register_native`. The ACP
9
9
  > backends are Claude and Snowflake Cortex Code. Antigravity is a separate shipped
10
- > native-push citizen lane; Codex has delivery-probe evidence but no managed citizen
11
- > lane. Retired v1 verbs and bridge implementations belong only in CHANGELOG/git.
10
+ > native-push citizen lane. Codex is a native-push citizen supported in 0.21.0, whose birth hook is
11
+ > operator-owned and gated on one vendor trust receipt, with strict request-scoped identity and
12
+ > visible fresh; it remains outside ACP and has no resume.
13
+ > The amended candidate carries its own full acceptance: standalone LIVE 48 assertions exit 0,
14
+ > `check:full` exit 0, and the LIVE release gate `--cut` at MUST 24/0/0 with `check-gate-qualification`
15
+ > killing 475/475 and both purity axes green. Earlier focused, 460-mutant, and 57-assertion receipts
16
+ > stay their own pre-amendment axes and are not promoted across it.
17
+ > Retired v1 verbs and bridge implementations belong only in CHANGELOG/git.
12
18
 
13
19
  This is a working protocol, not a metrics ledger. Per-run counts, digests, and release
14
20
  chronology belong in [BASELINE.md](./BASELINE.md), CHANGELOG, and release artifacts.
@@ -64,6 +70,27 @@ Verification here is not a benchmark. In production we exchange short turns and
64
70
  >
65
71
  > The aggregate release gate does not own a live agy conversation id, so agy's real native-push round trip is a separate acceptance axis: three fail-loud doctors plus `LIVE=1 AGY_CONVERSATION_ID=<id> ./run.sh smoke-agy-native-push-live`, followed by a fresh-conversation sender/reply check after package install. Its deterministic install/sender gates are already inside `pnpm run check:full`; do not misreport the aggregate gate as live agy evidence. **Cost fence:** the agy conversation this smoke drives runs on a free account, so open it on `gemini-3.6-flash` — never a Pro tier. The model is the operator's choice at conversation-open time; entwurf never selects it, and no assertion reads it (see the shipped-lane note: model display is not part of the agy contract).
66
72
  >
73
+ > Aggregate release-gate does not own a loaded Codex thread. Codex native-push remains an
74
+ > on-demand host axis. The final first-admission acceptance passed on 2026-09-12 with the
75
+ > explicit home topology; `DELIVERY.md` owns its 57-assertion receipt and digest. Re-runs start
76
+ > from a tmux session other than `codex`; the operator-owned
77
+ > app-server and supported Codex TUIs sit in the existing exact `codex` session; omitted Codex
78
+ > placement resolves there; and the Codex-opened Pi stays there. Exact `rust-v0.153.4` source
79
+ > exposes no arbitrary request→attached-TUI-seat value, so that wider topology is unsupported
80
+ > and unclaimed rather than a blocker to the explicit home.
81
+ >
82
+ > The A cell takes no inferred server or models:
83
+ > `LIVE=1 ENTWURF_CODEX_APP_SERVER_PID=<existing-app-server-pid>
84
+ > ENTWURF_CODEX_FRESH_MODEL=<codex-model>
85
+ > ENTWURF_CODEX_FRESH_PI_MODEL=<pi-model> ./run.sh smoke-codex-fresh-live`.
86
+ >
87
+ > The command name is not proof. The first leg must be a real record-backed visible Pi
88
+ > citizen outside the Codex home opening Codex; if a fixture or self-fetch receipt collector
89
+ > is what OPENS Codex, the gate does not satisfy the topology. A fixture that only COLLECTS
90
+ > receipts beside that real Pi leg is not the disqualifier — substituting for the leg is (see
91
+ > `docs/adding-a-harness.md`, the same wording). The receipt must separately name the initial
92
+ > Pi, app-server, fresh Codex, and outbound Pi coordinates.
93
+ >
67
94
  > Authoritative per-cut counts and digests live in BASELINE/CHANGELOG, not inline
68
95
  > here; embedding them in the protocol makes a correct guide stale after every cut.
69
96
 
@@ -81,6 +108,7 @@ Do not collapse source, package, fixture, and native-host evidence into one “g
81
108
  | LIVE runtime | `LIVE=1 ./run.sh release-gate <scratch> --cut` plus any shipped on-demand backend axis | `--cut` enforces `SKIP=0`; a red wired gate blocks the cut. |
82
109
  | Native Claude host | installed strict doctor against a new real session | Missing live join is `NOT CERTIFIED`, not a fixture PASS. |
83
110
  | Native agy host | three doctors plus conversation-id-gated native-push round trip | Aggregate release-gate does not own an agy conversation id. |
111
+ | Native Codex host | explicit-home first admission accepted 2026-09-12 | Operator-owned birth/MCP/statusline atoms and the on-demand four-coordinate chain are measured on Linux at Codex 0.153.4: initial Pi outside the existing exact `codex` home; app-server, omitted-placement fresh Codex, and outbound Pi inside it; exact callbacks and addressed delivery both ways. `DELIVERY.md` owns the 57-assertion receipt and digest. Arbitrary attached-TUI placement is unsupported and unclaimed. |
84
112
 
85
113
  The repo-local `entwurf-release` skill owns the `land → prepare → make → publish`
86
114
  state machine. Each mode is a separate GLG authorization. Preserve one candidate,
@@ -143,8 +171,9 @@ The goal is not merely "invoke Claude Code." We want:
143
171
  4. source stable-bin exposure — including certified `entwurf` → this checkout's `run.sh`, the managed runtime Copilot fresh resolves; helper units are attempted independently and a foreign helper is a named FAIL
144
172
  5. agy bridge + exact permission + statusline + `PreInvocation` hook — only when `agy` is on PATH; each adapter is idempotent and independently doctorable
145
173
  6. Copilot four-unit composition (birth → MCP → receiver → visible footer) — only when `copilot` is on PATH (#86 C3b); the units run independently, each keeps its package-owned install-state and inverse, and a failed unit is a named component FAIL. The explicit `install-copilot-*`/`uninstall-copilot-*` surfaces remain the per-unit repair and inverse path
146
- 7. `entwurf-bridge` install smoke (`validate_entwurf_bridge`)
147
- 8. computed summary — per-component PASS/SKIP/FAIL; any detected-integration FAIL makes the whole command exit nonzero while valid components stay installed
174
+ 7. Codex composition — only when `codex` is on PATH: setup publishes all three operator-owned atoms and stays non-green until the vendor's trust receipt for the birth declaration exists; it never invokes sudo, never writes `[hooks.state]`, and never starts the app-server or creates the `codex` tmux home. Its setup cells prove the exact `env_vars` boundary, independent component outcomes, idempotence, the non-green-until-trusted verdict with its named operator repair, and that setup never writes the receipt itself. First-admission qualification follows the separate explicit-home LIVE; unrestricted attached-TUI placement is outside the claim.
175
+ 8. `entwurf-bridge` install smoke (`validate_entwurf_bridge`)
176
+ 9. computed summary — per-component PASS/SKIP/FAIL; any detected-integration FAIL makes the whole command exit nonzero while valid components stay installed
148
177
 
149
178
  ```bash
150
179
  git clone https://github.com/junghan0611/entwurf /path/to/entwurf && cd $_
@@ -300,9 +329,10 @@ The minimum passing bar:
300
329
  5. **Honest self-recognition:** the bridged model identifies its actual harness/backend, lists `entwurf-bridge` as the single MCP server with its seven current tools, and presents a backend-native (not normalized) tool surface.
301
330
  6. **Carrier separation honored:** engraving vs pi-context-augment kept distinct (§1A.0); no bridge-identity narrative attributed to the engraving carrier.
302
331
  7. **agy shipped lane accepted:** all three agy doctors are green; automatic birth/statusline/sender identity and same-gid native-push reply are confirmed in a fresh conversation. `agentId=meta-session/antigravity` is correct; model display is not part of that contract. Same-pid concurrent conversation invocation is not claimed.
303
- 8. **Boundary preservation across backends/machines:** for every shipped or explicitly probed backend, regardless of install path or host, no cross-backend tool-surface contamination and no confabulation about pi internals.
304
- 9. **Hygiene:** no orphan ACP children; no unexpected persisted session garbage (a turn-scoped `cwd:` fallback is never a persisted reuse).
305
- 10. **New-harness admission closed:** a release that introduces a native harness admitted under the #82 contract has that harness in `FRESH_CALL_BACKENDS` on all three public surfaces, with `check-harness-admission-parity` green and its clause 7 visible-fresh LIVE step green in the MUST tier. That release also owes the cross-harness leg the same release stop names — two dispatch receipts, an existing citizen's live turn delivered into the new citizen and the new citizen's live turn delivered into an existing one — recorded in `DELIVERY.md`; its deterministic half is an owed follow-up, so until that gate lands this half of the condition is prose and is judged by the recorded receipts. Partial evidence is a branch state; an `unsupported` note in `DELIVERY.md` is a description, never a permit.
332
+ 8. **Codex first-admission home topology accepted:** its birth/MCP/status-line doctors are green and the on-demand real visible `Pi Codex Pi` cell records four coordinates: initial Pi in a session other than the exact existing `codex` home; operator-owned app-server, omitted-placement fresh Codex, and Codex-opened outbound Pi in that home; exact callbacks and addressed delivery succeed both ways. Missing home/app-server rejects without mutation. Entwurf neither creates nor supervises them, infers attached-TUI panes, nor claims unrestricted client placement. Codex remains outside ACP and has no resume claim.
333
+ 9. **Boundary preservation across backends/machines:** for every shipped or explicitly probed backend, regardless of install path or host, no cross-backend tool-surface contamination and no confabulation about pi internals.
334
+ 10. **Hygiene:** no orphan ACP children; no unexpected persisted session garbage (a turn-scoped `cwd:` fallback is never a persisted reuse).
335
+ 11. **New-harness admission closed:** a release that introduces a native harness admitted under the #82 contract has that harness in `FRESH_CALL_BACKENDS` on all three public surfaces, with `check-harness-admission-parity` green and its clause 7 visible-fresh LIVE step green in the MUST tier. That release also owes the cross-harness leg the same release stop names — two dispatch receipts, an existing citizen's live turn delivered into the new citizen and the new citizen's live turn delivered into an existing one — recorded in `DELIVERY.md`; its deterministic half is an owed follow-up, so until that gate lands this half of the condition is prose and is judged by the recorded receipts. Partial evidence is a branch state; an `unsupported` note in `DELIVERY.md` never weakens this stop.
306
336
 
307
337
  Passing establishes a **release verification floor**, not an 8-hour/day operational guarantee. The floor says: gates hold, the agent honestly recognizes its environment, no tool surface is normalized away, no identity leaks, no orphans. It does **not** say a real-day workload (50–100+ turns, tool bursts, partial MCP failures, auth/version drift) survives — that needs L3–L5 evidence (appendix).
308
338
 
@@ -92,10 +92,10 @@ undifferentiated "supported" column is what let a Claude PASS read as if it also
92
92
 
93
93
  | Surface | Declaration | Class | What a green actually says |
94
94
  |---|---|---|---|
95
- | Entwurf package | `0.17.0` | shipped baseline | the package contract these rows belong to |
95
+ | Entwurf package | `0.20.0` | shipped baseline | the package contract these rows belong to |
96
96
  | pi runtime | devDep exact `0.85.1`, peer `>=0.85.1 <0.86` | **exact** oracle + **closed range** | built and certified against 0.85.1; hosts inside the range are accepted, and the ceiling moves only on measurement |
97
97
  | ACP wire SDK | `@agentclientprotocol/sdk 1.4.0` | **exact** | the shared wire oracle both adapters speak |
98
- | Claude ACP adapter | `@agentclientprotocol/claude-agent-acp 0.75.1` | **exact**, bundled | the adapter we ship and certify; resolved before any PATH fallback |
98
+ | Claude ACP adapter | `@agentclientprotocol/claude-agent-acp 0.76.0` | **exact**, bundled | the adapter we ship and certify; resolved before any PATH fallback |
99
99
  | Claude Agent SDK | `0.3.257` (transitive) | **exact** oracle | the runtime risk surface behind the adapter |
100
100
  | Anthropic SDK | `0.100.1` | **exact**, peer-resolution only | satisfies the Agent SDK peer floor (0.93.0+); never an API client here (gate L4) |
101
101
  | Claude Code runtime | `>=2.1.217` (`entwurf.claudeCodeFloor`) | **floor** | below it, hook args are silently dropped; entwurf enforces this itself |
@@ -151,6 +151,23 @@ different reasons, and collapsing them would hide a real risk**:
151
151
  entwurf passes that flag nowhere (repo grep, 0 hits). They are
152
152
  unreachable only because the common loop never invokes them (nor `logout`). Nothing upstream
153
153
  enforces that; it is our own call-site discipline, and it stops holding the moment we use one.
154
+ - **Advertised but never called — 0.76.0 adds exactly one.** Upstream #1111 puts a
155
+ `recommendedValue` on AIR config options, and it is **opt-in**: the adapter turns it on only
156
+ when `initialize`'s `clientCapabilities._meta.jetbrains.air.capabilities` names
157
+ `recommendedValue`. entwurf sends `clientCapabilities: {}` (`backend.ts:1758`), so
158
+ `useRecommendedValue` is false on every branch. The label normalization, the `default` row and
159
+ the effort selector that ride it are unreachable for a second, independent reason as well:
160
+ `[측정 2026-09-10]` `git grep -c configOptions pi-extensions/lib/acp/` is **0**, and entwurf
161
+ has never set `EFFORT_CONFIG_ID`, which is the other gate on the SDK-side `applyFlagSettings`.
162
+ The rest of 0.76.0 is one refactor: `resolveModelPreference` moved into a new
163
+ `dist/session-model.*` (diff is comments and formatting), `dist/session-effort.*` is new, and
164
+ `setSessionConfigOption` — our only model-forcing wire call — is byte-identical across
165
+ `v0.75.1..v0.76.0`.
166
+ - **One 0.76.0 change has a reachable SHAPE and is still inert for us.** When
167
+ `configuredSettings` is a STRING PATH, its `readFile` + `JSON.parse` moved out of
168
+ `resolvedProvider` and up onto the unconditional `session/new` path. entwurf passes `settings`
169
+ as an OBJECT (`tool-surface.ts:153`), so the branch is never taken. That makes "entwurf never
170
+ hands `settings` as a path" a contract rather than an accident; it is stated at that call site.
154
171
  - **The one 0.73.0 → 0.75.1 change that DOES reach us:** context compaction is now surfaced as a
155
172
  synthetic ACP tool lifecycle (0.75.0, #991) — a `tool_call` with `kind: "think"`, title
156
173
  `Compact conversation`, and `_meta.contextCompaction` schema v1 — where it used to arrive as
@@ -275,23 +292,23 @@ caller-session `_meta`, and cross-machine certification.
275
292
 
276
293
  A backend can return `newSession` before its declared MCP server is callable. This was
277
294
  observed intermittently on the Claude rail and directly on Cortex's private `mcp.json`
278
- path. Neither `claude-agent-acp` 0.75.1 nor the Cortex landing adds a client-side
295
+ path. Neither `claude-agent-acp` 0.76.0 nor the Cortex landing adds a client-side
279
296
  readiness fence over a session's declared MCP servers, and entwurf's common loop
280
297
  calls `mcpServerStatus()` nowhere.
281
- (Re-measured at the 0.73.0 → 0.75.1 bump, not inherited from the previous one and the
282
- 0.70.0 → 0.73.0 argument is not reused either. `mcpServerStatus` call sites in
283
- `src/acp-agent.ts` are **2 at both v0.73.0 and v0.75.1** `[측정 2026-09-06, git grep -c]`;
284
- they first appeared in 0.71.0 via `0cbbaf3` (MCP OAuth, LLM-25012), so the ADAPTER calls it
285
- where it once did not. Both were re-read at `v0.75.1 src/acp-agent.ts:1736` and `:1829`
286
- (the 0.73.0 coordinates were `:1618` and `:1711`): the first sits inside
298
+ (Re-measured at the 0.75.1 → 0.76.0 bump, not inherited the previous bump's argument is
299
+ not reused, the way the 0.73.0 → 0.75.1 entry did not reuse 0.70.0 0.73.0's.
300
+ `mcpServerStatus` call sites in `src/acp-agent.ts` are **2 at v0.75.1 and 2 at v0.76.0**
301
+ `[측정 2026-09-10, upstream v0.76.0/src/acp-agent.ts read directly, grep -n]`; they first
302
+ appeared in 0.71.0 via `0cbbaf3` (MCP OAuth, LLM-25012), so the ADAPTER calls it where it
303
+ once did not. Both were re-read at `v0.76.0 src/acp-agent.ts:1762` and `:1855`
304
+ (v0.75.1: `:1736` / `:1829`; v0.73.0: `:1618` / `:1711`): the first sits inside
287
305
  `authenticateMcpServers` behind `supportsMcpOAuth(query)` and skips every status that is not
288
306
  `needs-auth`; the second polls a SINGLE named server to `connected` under an OAuth deadline.
289
307
  Neither waits on every declared server before `newSession` returns. That is an auth
290
- handshake, not a readiness fence, so the boundary below is unchanged. The other reachable-surface findings also re-measured:
291
- AIR typed failures and the AIR file-change report stay capability-gated and
292
- unadvertised by entwurf; `providers/set` / `providers/disable` stay advertised
293
- unconditionally and uncalled; native subagents, async tasks, session forks, session
294
- titles, permission-mode kinds and clear-context planning are all new-but-uncalled.
308
+ handshake, not a readiness fence, so the boundary below is unchanged. The surrounding 200
309
+ lines are byte-identical and the region moved +26; the whole 0.76.0 delta is one refactor
310
+ plus one opt-in AIR extension that `clientCapabilities: {}` never enables, so the other
311
+ reachable-surface findings stand as re-measured at the previous bump.
295
312
  This bump changes no readiness behavior and closes no part of #72.)
296
313
 
297
314
  ### 11-7-a/b. Instrument and first measurement
@@ -384,5 +401,4 @@ Receipt, limits and the `completed`-branch gap: `scripts/raw-acp-compaction-meas
384
401
 
385
402
  - causal MCP-readiness diagnosis and, only with proof, a backend-invariant fence;
386
403
  - broader installed-host and cross-machine Cortex evidence;
387
- - any future Codex managed native-citizen lane—separate from ACP; 0.14.0 does not ship one;
388
404
  - persisted ACP resume/load, which is not implemented by today's in-memory reuse.