gentle-pi 4.1.0 → 4.3.0

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 (249) hide show
  1. package/README.md +352 -128
  2. package/assets/agents/gentle-ai-explore.md +1 -0
  3. package/assets/agents/gentle-ai-verify.md +6 -4
  4. package/assets/agents/gentle-ai-worker.md +7 -1
  5. package/assets/orchestrator-delegation-reference.md +163 -0
  6. package/assets/orchestrator-delegation.md +25 -176
  7. package/assets/orchestrator-tracking.md +1 -1
  8. package/assets/orchestrator-verification.md +14 -8
  9. package/assets/orchestrator-writer.md +5 -1
  10. package/assets/orchestrator.md +9 -9
  11. package/bin/engram-guard +1 -1
  12. package/bin/gentle-shell.mjs +188 -35
  13. package/contracts/review-provider-contract-mirror/provider-contract.lock.json +8 -7
  14. package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/README.md +7 -0
  15. package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/manifest.json +11 -2
  16. package/contracts/review-provider-contract-mirror/v1.4.0/bundle/routing/pi.md +92 -0
  17. package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/generated/provider-capabilities.baseline.json +8 -1
  18. package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/generated/provider-roles.baseline.json +1 -1
  19. package/docs/assets/diagrams/agent-orchestration.svg +54 -15
  20. package/docs/assets/diagrams/gentleman-workflow.svg +69 -11
  21. package/docs/assets/diagrams/measured-routing.svg +49 -0
  22. package/docs/assets/diagrams/native-review.svg +52 -12
  23. package/docs/assets/diagrams/odd-workflow.svg +128 -66
  24. package/docs/assets/diagrams/permission-modes.svg +51 -0
  25. package/docs/assets/features/audio-notifications.png +0 -0
  26. package/docs/assets/features/command-palette.png +0 -0
  27. package/docs/auto-permission-answers.md +25 -0
  28. package/docs/auto-permission-evidence.md +39 -0
  29. package/docs/bundled-install.md +40 -8
  30. package/docs/delegated-verification.md +10 -8
  31. package/docs/gentle-agents-activity.md +11 -5
  32. package/docs/gentle-shell-uninstall.md +8 -1
  33. package/docs/gentle-shell.md +20 -12
  34. package/docs/install-wizard.md +17 -10
  35. package/docs/permission-modes.md +51 -6
  36. package/docs/readme-reference.md +155 -31
  37. package/docs/session-profile-format.md +39 -0
  38. package/docs/yolo-mode.md +21 -7
  39. package/extensions/ask-user-choice.ts +4 -0
  40. package/extensions/ask-user-question.ts +6 -1
  41. package/extensions/codegraph-tools.ts +68 -6
  42. package/extensions/gentle-agents.ts +146 -28
  43. package/extensions/gentle-ai.ts +380 -68
  44. package/extensions/gentle-herdr-activity.ts +2 -3
  45. package/extensions/gentle-shell.ts +53 -43
  46. package/extensions/gentle-todo.ts +62 -18
  47. package/extensions/quiet-tools.ts +1 -1
  48. package/extensions/session-scratch.ts +365 -0
  49. package/extensions/startup-banner.ts +2 -3
  50. package/lib/agent-home.ts +8 -3
  51. package/lib/agents-protocol.ts +3 -0
  52. package/lib/agents-runner.ts +55 -6
  53. package/lib/agents-view.ts +3 -1
  54. package/lib/auto-permission-answers.ts +133 -0
  55. package/lib/auto-permission-classifier.ts +215 -36
  56. package/lib/auto-permission-evidence.ts +88 -0
  57. package/lib/auto-permission-memory.ts +69 -0
  58. package/lib/auto-permission-runtime.ts +251 -35
  59. package/lib/auto-permission-tool-context.ts +61 -0
  60. package/lib/bounded-writer-admission.ts +51 -0
  61. package/lib/command-palette-catalog.ts +0 -4
  62. package/lib/commit-slice-signal.ts +100 -0
  63. package/lib/context-backstop.ts +84 -0
  64. package/lib/delivered-test-lint.ts +80 -0
  65. package/lib/destructive-command-guard.ts +215 -28
  66. package/lib/gauge-theme.ts +146 -0
  67. package/lib/gentle-ai-binary.ts +28 -7
  68. package/lib/gentle-shell-bundled.ts +62 -8
  69. package/lib/gentle-shell-import-pi.ts +233 -0
  70. package/lib/gentle-shell-launcher.ts +146 -12
  71. package/lib/gentle-shell-uninstall.ts +18 -1
  72. package/lib/herdr-activity.ts +24 -0
  73. package/lib/model-routing-authority.ts +6 -1
  74. package/lib/native-review-cli.ts +9 -1
  75. package/lib/orchestrator-discovery.ts +5 -1
  76. package/lib/orchestrator-presence.ts +9 -0
  77. package/lib/orchestrator-scope.ts +21 -3
  78. package/lib/provider-contract-bundle.ts +99 -53
  79. package/lib/review-candidate-view-owner.ts +29 -19
  80. package/lib/review-candidate-view.ts +36 -2
  81. package/lib/review-risk-assessment.ts +43 -38
  82. package/lib/review-session-standing-permission.ts +3 -1
  83. package/lib/selection-engine.ts +4 -1
  84. package/lib/session-profile-freeze.ts +40 -16
  85. package/lib/session-profile-persistence.ts +187 -0
  86. package/lib/session-scratch.ts +474 -0
  87. package/lib/shell-bar.ts +25 -25
  88. package/lib/shell-changes-view.ts +80 -13
  89. package/lib/shell-changes.ts +2 -0
  90. package/lib/shell-gauge.ts +49 -6
  91. package/lib/shell-prompt.ts +6 -4
  92. package/lib/shell-todo.ts +39 -8
  93. package/lib/shell-usage-view.ts +5 -1
  94. package/lib/shell-usage.ts +7 -7
  95. package/lib/verifier-admission.ts +29 -0
  96. package/lib/vim-editor-adapter.ts +22 -4
  97. package/lib/yolo-temp-scratch.ts +69 -0
  98. package/package.json +5 -6
  99. package/runtime/gentle-ai-binary.mjs +28 -7
  100. package/runtime/gentle-shell-bundled.mjs +61 -7
  101. package/runtime/gentle-shell-import-pi.mjs +234 -0
  102. package/runtime/gentle-shell-launcher.mjs +144 -10
  103. package/runtime/gentle-shell-uninstall.mjs +18 -1
  104. package/runtime/native-review-cli.mjs +8 -0
  105. package/runtime/review-risk-assessment.mjs +43 -38
  106. package/scripts/build-runtime-modules.mjs +1 -0
  107. package/scripts/bundled-engram-migration.mjs +125 -0
  108. package/scripts/gentle-ai-installer.mjs +10 -10
  109. package/scripts/install-wizard-preview.mjs +2 -2
  110. package/scripts/installer-downloads.mjs +7 -7
  111. package/scripts/installer-preflight.mjs +1 -1
  112. package/scripts/installer-server.mjs +14 -4
  113. package/scripts/main-channel.mjs +3 -2
  114. package/scripts/pi-update.mjs +223 -0
  115. package/scripts/run-test-suite.mjs +5 -0
  116. package/scripts/verify-package-files.mjs +19 -13
  117. package/skills/gentle-ai/SKILL.md +3 -3
  118. package/skills/work-unit-commits/SKILL.md +5 -3
  119. package/tests/agent-risk-escalation.test.ts +37 -10
  120. package/tests/agents-runner.test.ts +260 -3
  121. package/tests/append-system-prompt-route.test.ts +18 -1
  122. package/tests/auto-mode-human-answers.test.ts +68 -0
  123. package/tests/auto-mode-invariants.test.ts +23 -5
  124. package/tests/auto-mode-sdk.test.ts +29 -2
  125. package/tests/auto-permission-answers.test.ts +170 -0
  126. package/tests/auto-permission-classifier.test.ts +353 -2
  127. package/tests/auto-permission-compound.test.ts +24 -0
  128. package/tests/auto-permission-evidence.test.ts +88 -0
  129. package/tests/auto-permission-memory.test.ts +46 -0
  130. package/tests/auto-permission-public-regressions.test.ts +111 -0
  131. package/tests/auto-permission-runtime.test.ts +270 -12
  132. package/tests/autonomous-guard.test.ts +208 -16
  133. package/tests/bounded-writer-admission.test.ts +11 -3
  134. package/tests/bundled-install.test.ts +3 -3
  135. package/tests/card-content-policy.test.ts +19 -9
  136. package/tests/child-safety.test.ts +39 -0
  137. package/tests/codegraph-tools.test.ts +123 -2
  138. package/tests/command-palette.test.ts +14 -4
  139. package/tests/commit-slice-signal.test.ts +162 -0
  140. package/tests/context-backstop.test.ts +159 -0
  141. package/tests/delegate-for-reason-contract.test.ts +91 -24
  142. package/tests/delivered-test-lint.test.ts +105 -0
  143. package/tests/destructive-command-guard.test.ts +85 -1
  144. package/tests/engram-safety.test.ts +35 -7
  145. package/tests/fixtures/primary-session-prompt/gentleman.golden.md +143 -0
  146. package/tests/fixtures/provider-contract-bundle/v1.4.0/README.md +29 -0
  147. package/tests/fixtures/provider-contract-bundle/v1.4.0/manifest.json +83 -0
  148. package/tests/fixtures/provider-contract-bundle/v1.4.0/orchestration/pi.md +70 -0
  149. package/tests/fixtures/provider-contract-bundle/v1.4.0/routing/pi.md +92 -0
  150. package/tests/fixtures/provider-contract-bundle/v1.4.0/schemas/lens.schema.json +16 -0
  151. package/tests/fixtures/provider-contract-bundle/v1.4.0/schemas/refuter.schema.json +1 -0
  152. package/tests/fixtures/provider-contract-bundle/v1.4.0/schemas/targeted-validator.schema.json +1 -0
  153. package/tests/fixtures/provider-contract-bundle/v1.4.0/vectors/lens.json +1 -0
  154. package/tests/fixtures/provider-contract-bundle/v1.4.0/vectors/refuter.json +1 -0
  155. package/tests/fixtures/provider-contract-bundle/v1.4.0/vectors/targeted-validator.json +1 -0
  156. package/tests/gauge-theme.test.ts +120 -0
  157. package/tests/gentle-agents.test.ts +443 -8
  158. package/tests/gentle-ai-binary.test.ts +1 -1
  159. package/tests/gentle-ai-child-guards.test.ts +61 -0
  160. package/tests/gentle-ai-doctor-herdr.test.ts +94 -0
  161. package/tests/gentle-ai-installer.test.ts +51 -51
  162. package/tests/gentle-shell-bin.test.ts +180 -17
  163. package/tests/gentle-shell-bundled.test.ts +235 -38
  164. package/tests/gentle-shell-import-pi.test.ts +262 -0
  165. package/tests/gentle-shell-launcher.test.ts +97 -13
  166. package/tests/gentle-shell-uninstall.test.ts +30 -0
  167. package/tests/gentle-shell.test.ts +113 -14
  168. package/tests/gentle-todo.test.ts +258 -27
  169. package/tests/headless-delivery-question.test.ts +58 -0
  170. package/tests/herdr-activity.test.ts +34 -1
  171. package/tests/install-wizard.test.ts +2 -2
  172. package/tests/installer-posix-bootstrap.test.ts +8 -8
  173. package/tests/installer-preflight.test.ts +3 -3
  174. package/tests/installer-probes.test.ts +8 -8
  175. package/tests/installer-runner.test.ts +14 -13
  176. package/tests/installer-server.test.ts +13 -5
  177. package/tests/installer-windows-bootstrap.test.ts +33 -4
  178. package/tests/main-channel-upgrade.test.ts +7 -1
  179. package/tests/native-review-capability-contract.test.ts +7 -1
  180. package/tests/native-review-parity.test.ts +22 -1
  181. package/tests/no-self-pause-contract.test.ts +5 -2
  182. package/tests/odd-integration.test.ts +5 -1
  183. package/tests/odd-routing-canonical-ratchet.test.ts +127 -162
  184. package/tests/odd-routing-contract.test.ts +115 -30
  185. package/tests/odd-runtime-delegation-gate.test.ts +4 -1
  186. package/tests/orchestrator-budget.test.ts +19 -8
  187. package/tests/orchestrator-discovery.test.ts +29 -0
  188. package/tests/orchestrator-rdd-ownership.test.ts +3 -1
  189. package/tests/orchestrator-scope.test.ts +66 -0
  190. package/tests/orchestrator-status-publication-contract.test.ts +12 -8
  191. package/tests/package-manifest.test.ts +28 -24
  192. package/tests/pi-update-launcher.test.ts +106 -0
  193. package/tests/pi-update.test.ts +186 -0
  194. package/tests/primary-session-prompt.test.ts +101 -0
  195. package/tests/provider-contract-bundle.test.ts +118 -0
  196. package/tests/provider-contract-mirror.test.ts +24 -0
  197. package/tests/quiet-tool-rendering.test.ts +23 -0
  198. package/tests/rdd-aware-verification-contract.test.ts +29 -11
  199. package/tests/rdd-status-line.test.ts +131 -0
  200. package/tests/review-agent-end-preflight.test.ts +6 -0
  201. package/tests/review-candidate-view.test.ts +49 -0
  202. package/tests/review-contract-prompt.test.ts +5 -2
  203. package/tests/review-controller-native-routing.test.ts +14 -1
  204. package/tests/review-controller.test.ts +1 -1
  205. package/tests/review-host-relay-routing.test.ts +59 -2
  206. package/tests/review-loadout-rdd-off-sdk.test.ts +237 -0
  207. package/tests/review-owner-identity.test.ts +132 -0
  208. package/tests/review-risk-assessment.test.ts +163 -76
  209. package/tests/review-session-standing-permission-controller.test.ts +14 -1
  210. package/tests/review-session-standing-permission.test.ts +27 -1
  211. package/tests/run-test-suite.test.ts +15 -0
  212. package/tests/runtime-harness.mjs +16 -6
  213. package/tests/selection-engine.test.ts +35 -5
  214. package/tests/session-changes-shell.test.ts +1 -1
  215. package/tests/session-profile-freeze.test.ts +37 -1
  216. package/tests/session-profile-persistence.test.ts +334 -0
  217. package/tests/session-scratch.test.ts +945 -0
  218. package/tests/session-subject-routing-contract.test.ts +2 -1
  219. package/tests/shell-bar.test.ts +7 -0
  220. package/tests/shell-changes-view.test.ts +137 -1
  221. package/tests/shell-changes.test.ts +18 -0
  222. package/tests/shell-gauge.test.ts +42 -0
  223. package/tests/shell-todo.test.ts +15 -12
  224. package/tests/small-task-inline-contract.test.ts +37 -0
  225. package/tests/standalone-config-home.test.ts +324 -0
  226. package/tests/startup-banner.test.ts +5 -4
  227. package/tests/startup-home-repo.test.ts +34 -0
  228. package/tests/support/orchestrator-modules.ts +1 -0
  229. package/tests/task-size-routing-contract.test.ts +69 -9
  230. package/tests/test-scratch-contract.test.ts +102 -0
  231. package/tests/test-strength-contract.test.ts +36 -1
  232. package/tests/tool-batching-contract.test.ts +64 -0
  233. package/tests/verification-proportionality.test.ts +46 -0
  234. package/tests/verifier-admission.test.ts +137 -0
  235. package/tests/verify-agent-contract.test.ts +8 -8
  236. package/tests/vim-editor-adapter.test.ts +41 -15
  237. package/tests/windows-hidden-processes.test.ts +5 -0
  238. package/tests/writer-edit-surface-scope.test.ts +3 -0
  239. package/tests/writer-feature-document-admission.test.ts +156 -0
  240. package/tests/yolo-mode.test.ts +24 -1
  241. package/tests/yolo-temp-scratch.test.ts +71 -0
  242. package/scripts/mirror-odd-routing.mjs +0 -242
  243. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/orchestration/pi.md +0 -0
  244. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/schemas/lens.schema.json +0 -0
  245. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/schemas/refuter.schema.json +0 -0
  246. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/schemas/targeted-validator.schema.json +0 -0
  247. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/vectors/lens.json +0 -0
  248. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/vectors/refuter.json +0 -0
  249. /package/contracts/review-provider-contract-mirror/{v1.3.0 → v1.4.0}/bundle/vectors/targeted-validator.json +0 -0
package/README.md CHANGED
@@ -32,8 +32,6 @@
32
32
 
33
33
  <p align="center">Your terminal can run an agent. Your workspace should help you lead it.<br><strong>gentle-shell is your coding agent, bringing your changes, tasks, and engineering workflow together—built for Pi.</strong></p>
34
34
 
35
- <p align="center"><sub>One workspace. A coding agent you direct. A workflow you can inspect.</sub></p>
36
-
37
35
  <div align="center">
38
36
  <h3>🎬 See it in action</h3>
39
37
 
@@ -44,67 +42,65 @@
44
42
  https://github.com/user-attachments/assets/fa5c0cfe-06e7-4c0d-bd6e-8ac7cb934339
45
43
 
46
44
 
47
- <p>Prefer Spanish subtitles?</p>
48
-
45
+ <details>
46
+ <summary><kbd>Watch</kbd> <strong>Prefer Spanish subtitles?</strong> — the same demo with Spanish captions</summary>
47
+
48
+ <br>
49
49
 
50
50
  https://github.com/user-attachments/assets/6d2bc422-a4dd-4ecf-a04b-fcd3bea7fea9
51
51
 
52
52
 
53
+ </details>
54
+
53
55
  </div>
54
56
 
55
- <p align="center"><strong>BUILT FOR PI</strong> &nbsp;·&nbsp; Coding-agent workspace &nbsp;·&nbsp; Focused agents &nbsp;·&nbsp; ODD</p>
57
+ <p align="center"><strong>BUILT FOR PI</strong> &nbsp;·&nbsp; Shaped by Gentle-AI &nbsp;·&nbsp; Coding-agent workspace &nbsp;·&nbsp; Focused agents &nbsp;·&nbsp; ODD</p>
56
58
 
57
59
  <p align="center">
58
- <a href="https://github.com/Gentleman-Programming/gentle-shell/stargazers"><strong>★ Star gentle-shell on GitHub</strong></a>
60
+ <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
59
61
  </p>
60
62
 
61
- <div align="center">
63
+ <a id="whats-new-in-v41"></a>
62
64
 
63
- <!--
64
- sealed_token is a GitHub fine-grained token encrypted against Star History's
65
- public key, so only the encrypted value is published here. It is required
66
- because GitHub restricted the stargazers API to a repository's admins and
67
- collaborators on 2026-06-30; without it the chart renders an error placeholder.
68
- Regenerate it at https://www.star-history.com/?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left
69
- -->
65
+ ## ✨ What's new in v4.1
70
66
 
71
- <a href="https://www.star-history.com/?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left">
72
- <picture>
73
- <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&theme=dark&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
74
- <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
75
- <img width="620" alt="Star History Chart" src="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
76
- </picture>
77
- </a>
78
-
79
- <br>
67
+ The [v4.1.0 release](https://github.com/Gentleman-Programming/gentle-shell/releases/tag/v4.1.0) makes Gentle Shell installable on a clean machine and pins Gentle AI v4.1.1:
80
68
 
81
- <sub>Built for Pi. Shaped by Gentle-AI.</sub>
82
-
83
- </div>
69
+ - **Clean-machine installers:** a browser wizard and double-click installers for macOS, Windows, and Linux install what is missing, update what you have, and ask once before changing anything. See [Get started](#get-started).
70
+ - **Bundled Gentle Shell:** a new installation keeps its exact Pi, pinned Gentle AI, Engram, and state in one prefix; existing global installs can migrate, the launcher offers updates, and `gentle-shell self-uninstall` removes it.
71
+ - **Background jobs and monitors:** `bash_background` wakes the agent once when a command exits, and `monitor` turns each output line of a long-running command into an event. Inspect them with `/gentle:jobs` ([Background jobs](#background-jobs)).
72
+ - **Contextual Auto permissions:** Auto decides with the conversation's context and image input, retries provider failures, and can be the saved default ([Permission modes](#permission-modes--normal-auto-or-yolo)).
73
+ - **Cross-session coordination:** sessions publish a recognizable subject, task aliases, repository scope, and a bounded work classification that other sessions can list, search, and consult with explicit cost permission.
74
+ - **Session-bound profiles:** the active profile binds to the parent session, and orchestrator and reviewer routing apply per session.
75
+ - **Delegate for a reason and review by risk:** see [Focused agents](#focused-agents--context-with-a-return-path), [ODD](#odd--the-everyday-workflow), and [Native review](#native-review--review-the-exact-change).
84
76
 
85
77
  <p align="center">
86
78
  <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
87
79
  </p>
88
80
 
89
- ## Features
81
+ <a id="features"></a>
90
82
 
91
- ---
83
+ <a id="how-it-works"></a>
92
84
 
93
- ### gentle-shell — Your coding agent, in the workspace you lead
85
+ ## 🧠 How it works
94
86
 
95
- <img width="100%" src="https://github.com/user-attachments/assets/5d9eefc2-7b2a-48f8-b212-1439834ce195" alt="gentle-shell running a live agent session: a header row with branch, model, and context gauge above the transcript, with status, changes, and todo cards in the right rail">
87
+ ### el Gentleman — Think before you build
96
88
 
97
- A bare terminal answers "what is the agent doing?" only with scrollback. gentle-shell turns your Pi session into a workspace: agent orchestration, live changes and runtime status, usage monitoring for supported provider accounts, and built-in diff views — so you lead the work instead of chasing it.
89
+ <img width="100%" src="docs/assets/diagrams/gentleman-workflow.svg" alt="el Gentleman workflow as six steps: you give the intent; el Gentleman clarifies the scope, picks a sensible next step, does the work, and gathers evidence people can review; the final decision stays yours">
98
90
 
99
- <p align="center"><sub>gentle-shell in action. Screenshot from <a href="https://raw.githubusercontent.com/Gentleman-Programming/gentle-ai/main/docs/assets/features/gentle-shell.png">Gentle-AI</a>.</sub></p>
91
+ Say what you need once, then keep moving. el Gentleman helps turn intent into clear scope, a sensible next step, and evidence people can review — without making every task feel like a process meeting.
100
92
 
101
- **[Docs →](docs/gentle-shell.md)**
93
+ **[Docs →](docs/readme-reference.md#organic-driven-development)**
102
94
 
103
95
  ---
104
96
 
105
- ### el Gentleman — Think before you build
97
+ ### ODD — The everyday workflow
106
98
 
107
- Say what you need once, then keep moving. el Gentleman helps turn intent into clear scope, a sensible next step, and evidence people can review — without making every task feel like a process meeting.
99
+ <img width="100%" src="docs/assets/diagrams/odd-workflow.svg" alt="Organic Driven Development as seven numbered steps in one left-to-right flow: Authorize, Explore, and Resolve uncertainty to understand; Classify forks so small work stays light while substantial work gets step five, Track, with one feature document; both paths converge on Implement task by task and then Close with verified proof, above a band noting that one feature document mirrored in Engram lets work resume across sessions">
100
+
101
+ **Organic Driven Development (ODD)** is the everyday path: the agent explores before changing anything, clarifies only real decisions, and keeps small understood work small. Substantial, authorized work gets one recoverable feature document — mirrored in memory when available — so progress, evidence, and the next step survive an interruption; behavior changes are test-first by default when a runnable test applies.
102
+
103
+ Tasks are sized by understanding, risk, and resumability: work you could finish from the request and `git diff` alone runs inline, with no subagents and no task document. The feature document quotes your spec verbatim, and every handoff points at it instead of paraphrasing it, so exact strings, error messages, and examples reach the workers intact.
108
104
 
109
105
  **[Docs →](docs/readme-reference.md#organic-driven-development)**
110
106
 
@@ -112,41 +108,81 @@ Say what you need once, then keep moving. el Gentleman helps turn intent into cl
112
108
 
113
109
  ### Focused agents — Context with a return path
114
110
 
115
- <img width="100%" src="docs/assets/diagrams/agent-orchestration.svg" alt="Diagram of one parent session directing bounded map, implementation, and verification work and receiving evidence back">
111
+ <img width="100%" src="docs/assets/diagrams/agent-orchestration.svg" alt="Diagram of one parent session sending tasks to a read-only explorer, parallel writers, and an independent verifier, each returning its observed result to the parent, which stays accountable">
116
112
 
117
113
  Bring in help without losing the thread. Focused package-owned Pi agents can map a codebase, implement a bounded change, or verify it, while one parent stays accountable for the scope, the decisions, and the final summary.
118
114
 
115
+ The parent delegates for a reason, not for size. Understanding that outgrows one inline batch (about 3 calls and 10k tokens) gets one read-only explorer; two or more independent units of behavior, each big enough to outweigh a subagent's start-up and with disjoint edit surfaces, get parallel writers; a change that touches a named high-risk item gets an independent verifier, whatever model wrote it; a long session hands its next unit to a fresh worker. Without a named reason the parent works inline — file count and task size never decide the route.
116
+
119
117
  **[Docs →](docs/readme-reference.md#how-the-harness-decides-what-to-do)**
120
118
 
121
119
  ---
122
120
 
123
- ### ODD — The everyday workflow
121
+ ### Measured — What proportional routing changed
124
122
 
125
- <img width="100%" src="docs/assets/diagrams/odd-workflow.svg" alt="Organic Driven Development as seven numbered steps: Authorize, Explore, Resolve uncertainty, and Classify across the top row; Classify forks, so small understood work stays light while substantial work gets step five, Track, with one feature document; both paths converge on Implement task by task and then Close, above a dashed band marking that one feature document mirrored in Engram lets work resume across sessions">
123
+ <img width="100%" src="docs/assets/diagrams/measured-routing.svg" alt="Bar chart of wall-clock time before and after proportional routing: a 3-line bug fell from 335 to 77 seconds inline, a shared-logic bug from 479 to 276 seconds with one explorer, and a high-risk data migration went from 559 to 539 seconds but was accepted on the first round with one independent verifier">
126
124
 
127
- **Organic Driven Development (ODD)** is the everyday path: the agent explores before changing anything, clarifies only real decisions, and keeps small understood work small. Substantial, authorized work gets one recoverable feature document — mirrored in memory when available — so progress, evidence, and the next step survive an interruption; checks follow the configured TDD mode.
125
+ The routing rules came from benchmarks. These runs use a local benchmark with hidden acceptance suites; each cell is a single run (n=1), and cost is API-equivalent.
128
126
 
129
- **[Docs →](docs/readme-reference.md#organic-driven-development)**
127
+ | Scenario | Before | After |
128
+ | :--- | :--- | :--- |
129
+ | 3-line bug in one function | 335 s · 56 turns · $0.339 | **77 s · 10 turns · $0.105** — inline, no subagents |
130
+ | Bug rooted in shared logic | 479 s · $0.361 | **276 s · $0.207** — one explorer |
131
+ | Data migration (high risk) | 559 s · $0.403 · needed a feedback round | 539 s · $0.555 · **accepted on the first round** — one independent verifier |
132
+ | 73-line spec, 26 hidden checks | Paraphrased handoffs: 13 failing, still 23/26 after two feedback rounds | Spec passed by reference ([#1718](https://github.com/Gentleman-Programming/gentle-shell/pull/1718)) |
133
+
134
+ Against the same model in Codex CLI, with blind review of every solution, Gentle Shell was accepted on the first round in every cell and ranked higher on the medium (3 of 3 runs) and large tasks; it was close on the small one. The large task still took about three times longer (17.5–17.9 min vs 6.1 min) — that gap is open work. Sources: [#1723](https://github.com/Gentleman-Programming/gentle-shell/pull/1723), [#1794](https://github.com/Gentleman-Programming/gentle-shell/pull/1794), [#1713](https://github.com/Gentleman-Programming/gentle-shell/issues/1713).
130
135
 
131
136
  ---
132
137
 
133
138
  ### Native review — Review the exact change
134
139
 
135
- <img width="100%" src="docs/assets/diagrams/native-review.svg" alt="Diagram showing one frozen candidate passing through risk-scoped native review to an outcome, while human delivery choices stay separate">
140
+ <img width="100%" src="docs/assets/diagrams/native-review.svg" alt="Diagram of a frozen candidate passing through risk-scoped review to an evidence outcome, with a bounded correction path back to the candidate, while a separate boundary keeps the decision of what ships with you">
136
141
 
137
142
  Review the exact change, not a moving target. Native review keeps one candidate in view, returns risk-scoped evidence, and can surface a bounded correction path. You still decide what happens next in your repository.
138
143
 
144
+ Review is proportionate: an independent verifier runs only for a named high-risk item in every RDD state, a review is only due for high-risk work, and the agent forwards the 4R lenses that fit what it touched — with its reason — instead of paying for all four.
145
+
146
+ > ℹ️ **RDD is opt-in:** enable native receipt-driven development only through an explicit `/gentle:review-mode enable` decision. The `.git/gentle-ai/candidate-views` parent must sit on a filesystem that honors private POSIX modes (or equivalent Windows ACLs); WSL DrvFS mounts without metadata can reject START before lineage creation.
147
+
148
+ <details>
149
+ <summary><kbd>More</kbd> <strong>Reviewer transport</strong> — how review completions follow Pi's transport setting</summary>
150
+
151
+ <br>
152
+
139
153
  Reviewer completions inherit Pi's effective `transport` preference when an acknowledged individual or grouped capture executes. The host forwards a defined value unchanged, with one settings snapshot per group; it never forces SSE or changes model routing. An absent setting or an older Pi API without `getSettings()` preserves the provider default. This completion preference is separate from native `pi_host_relay` coordination, and providers that do not support it may ignore it.
140
154
 
155
+ </details>
156
+
157
+ <br>
158
+
141
159
  **[Docs →](docs/review-integration.md)**
142
160
 
161
+ <p align="center">
162
+ <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
163
+ </p>
164
+
165
+ <a id="your-workspace"></a>
166
+
167
+ ## 🖥️ Your workspace
168
+
169
+ ### gentle-shell — Your coding agent, in the workspace you lead
170
+
171
+ <img width="100%" src="https://github.com/user-attachments/assets/5d9eefc2-7b2a-48f8-b212-1439834ce195" alt="gentle-shell running a live agent session: a header row with branch, model, and context gauge above the transcript, with status, changes, and todo cards in the right rail">
172
+
173
+ A bare terminal answers "what is the agent doing?" only with scrollback. gentle-shell turns your Pi session into a workspace: agent orchestration, live changes and runtime status, usage monitoring for supported provider accounts, and built-in diff views — so you lead the work instead of chasing it.
174
+
175
+ <p align="center"><sub>gentle-shell in action. Screenshot from <a href="https://raw.githubusercontent.com/Gentleman-Programming/gentle-ai/main/docs/assets/features/gentle-shell.png">Gentle-AI</a>.</sub></p>
176
+
177
+ **[Docs →](docs/gentle-shell.md)**
178
+
143
179
  ---
144
180
 
145
181
  ### Gentle Changes — What your session left uncommitted
146
182
 
147
183
  <img width="100%" src="docs/assets/features/changes-view.png" alt="Gentle Changes viewer: worktree accordion with per-file status on the left, the diff against HEAD with line counts on the right, and a keyboard hint row">
148
184
 
149
- You should not have to run `git status` in every worktree to find out what your agents left behind. Gentle Changes shows the current Git state — working tree against HEAD, untracked files included — of your session's repository plus every worktree its subagents used, one group per root. Commit and it disappears; remove a worktree and its group goes with it. Nothing runs at startup; `alt+g` opens a two-pane viewer with per-file line counts and lazy diffs, and `o` drops the real file into your editor.
185
+ You should not have to run `git status` in every worktree to find out what your agents left behind. Gentle Changes shows the current Git state — working tree against HEAD, untracked files included — of your session's repository plus every worktree its subagents used, one group per root. Commit and it disappears; remove a worktree and its group goes with it. Nothing runs at startup; `alt+g` opens a full-screen two-pane viewer with per-file line counts and lazy diffs, `F` hides the file list to read the diff at full width, and `o` drops the real file into your editor.
150
186
 
151
187
  **[Docs →](docs/gentle-shell.md#browse-changes)**
152
188
 
@@ -172,39 +208,138 @@ Model, effort, and who does what should be choices, not accidents. Named profile
172
208
 
173
209
  ---
174
210
 
211
+ ### Command palette — Every command, one keystroke away
212
+
213
+ <img width="100%" src="docs/assets/features/command-palette.png" alt="Command palette open over the session: a search prompt above grouped entries for Configuration, Session, and Diagnostics, with alt+g and alt+a shortcuts on the right and a key hint row below">
214
+
215
+ Extension commands are only useful if you can find them. `alt+k` opens a curated, grouped palette — Configuration, Session, Diagnostics, and Skills — searchable by label, command name, or description, showing entries only when they are actually registered.
216
+
217
+ **[Docs →](docs/gentle-shell.md#command-palette)**
218
+
219
+ ---
220
+
221
+ ### Customize — Look, layout, and editor in one panel
222
+
223
+ `/gentle:customize` opens an interactive panel to set animation quality, startup banner rose, text logo and color, or choose an installed Pi theme.
224
+
225
+ <details>
226
+ <summary><kbd>More</kbd> <strong>Everything <code>/gentle:customize</code> controls</strong> — themes, status layout, Vim, history, and visual profiles</summary>
227
+
228
+ <br>
229
+
230
+ - **Themes:** Highlighting a theme previews its source palette without changing the active theme; press Enter or Space to apply it through Pi. If its source is unreadable, the preview is unavailable.
231
+ - **Status layout:** Status defaults to a right rail in fullscreen terminals at least 140 columns wide, and to a bottom bar otherwise. Choose right, bottom, or hidden (which removes both the rail and the bottom status bar at every width), move the fullscreen header below the input (below the 140-column breakpoint only one status row paints: a top header replaces the bottom bar, while with the header below the input the bottom bar alone carries the header's context, cost, and usage plus extension statuses), select comfortable/compact/minimal density, and toggle Changes, Agents, TODO, usage/cost, and model details independently. Layout changes and reset take effect immediately; banner changes appear on the next startup.
232
+ - **Vim:** The Editor category offers explicit Vim enable/disable controls for the global prompt preference; highlighting shows the persisted preference and effective prompt state without changing either. Enter or Space saves it and updates the live prompt; unsupported editors keep ordinary editing even when the saved preference is on. Vim is not included in visual profiles or visual reset.
233
+ - **History:** The History category turns prompt-history capture on or off; it is off by default, applies to the next prompt without a restart, and never deletes stored history. An explicit `GENTLE_PI_HISTORY_CAPTURE` value overrides the saved choice, and the panel marks that override (see [Prompt history](docs/prompt-history.md)).
234
+ - **Reset and profiles:** The panel can reset visual, banner, and animation settings to defaults. Press `p` for named visual profiles: `s` saves the current installed theme, banner, animation and layout; select a profile with ↑/↓, then use `r` to replace, `a` to apply, or `d` to delete. `z` clears only the profile catalog. Confirm destructive/apply actions with `y`, or cancel with any other key; Esc returns without applying a preview. Applying independent stores is not atomic: partial failures identify what changed.
235
+
236
+ </details>
237
+
238
+ <br>
239
+
240
+ **[Card style and content →](docs/gentle-shell.md#card-style)**
241
+
242
+ ---
243
+
244
+ ### Audio notifications — Opt-in sounds for success, errors, and attention
245
+
246
+ <img width="100%" src="docs/assets/features/audio-notifications.png" alt="Customize panel with the Notifications category selected: audio notifications on and unmuted, success, error, and attention tones, an Advanced row for per-event exceptions, and a preview line for the global switch">
247
+
248
+ Open `/gentle:customize` → **Notifications** (the card header reads **Audio notifications**) and configure audio directly in the same two-column card, with no nested menu. Audio starts off; RPC and children stay silent.
249
+
250
+ <details>
251
+ <summary><kbd>More</kbd> <strong>Controls and platform support</strong> — keys, custom sounds, and per-OS playback</summary>
252
+
253
+ <br>
254
+
255
+ - **Controls:** Enter toggles the master switch or cycles a type/event through silence/success/error/attention, `f` assigns a literal absolute local WAV/OGG/FLAC sound (≤2 MiB, ≤10 seconds) to the highlighted **Success**, **Error** or **Attention** type (each type keeps its own sound), and `p` previews its assigned sound. **Advanced** folds per-event exceptions; the type choices stay independent.
256
+
257
+ - **Quick answer (Linux/WSLg):** on plain Linux the installed package prefers the native WAV backend and needs no external player — it talks to a local PulseAudio/PipeWire-Pulse Unix socket and requires a running server with a default sink.
258
+ - Inside **WSL** the Windows system `SoundPlayer` is preferred instead: the validated WAV snapshot is mapped to a `\\wsl.localhost\<distro>\...` UNC path, so no Pulse server or RDP audio dependency is required; it does depend on the default `/mnt/c` automount and the standard `C:\Windows` root and otherwise falls back to the trusted Linux CLI before any playback.
259
+ - **Native Windows** WAV uses the same system `SoundPlayer` through the fixed `C:\Windows\...\powershell.exe` host. A read-only probe succeeded here, but that is not a claim of physically heard audio.
260
+ - OGG/FLAC keep the legacy CLI backend (`paplay`/`pw-play`/`aplay`) on Linux and WSL; native codecs are a future phase and Windows OGG/FLAC is unsupported. macOS keeps `afplay` for WAV and FLAC while an own CoreAudio phase is planned.
261
+
262
+ </details>
263
+
264
+ <br>
265
+
266
+ **[Usage, limits and verification →](docs/sound-notifications.md)**
267
+
268
+ ---
269
+
270
+ <a id="background-jobs"></a>
271
+
272
+ ### Background jobs — Inspect what runs in the background
273
+
274
+ Use `/gentle:jobs` to inspect this session's background commands and monitors. Running jobs appear first; each group is ordered newest first. Arrow keys select a job, `Tab` opens details on narrow terminals, `s` stops a running job, and `q` closes the modal.
275
+
276
+ The Status panel has a **Jobs** section with each running job's description. When the panel is hidden or unavailable in the compact/mobile layout, the top or bottom bar shows only the running count (`⧗ 1 job`). The count and descriptions disappear when no jobs remain active; `/gentle:jobs` retains completed jobs for inspection.
277
+
278
+ **[Docs →](docs/gentle-shell.md#background-jobs)**
279
+
280
+ <p align="center">
281
+ <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
282
+ </p>
283
+
284
+ <a id="control"></a>
285
+
286
+ ## 🛡️ Control
287
+
175
288
  ### Permission modes — Normal, Auto, or YOLO
176
289
 
177
- Choose `/gentle:mode auto` to check eligible actions with an independent, tool-free inference using the current session model. Auto covers primary file/shell/custom/MCP tools and nested calls through the SDK pipeline while retaining explicit restrictions, destructive guards and independent consent. It adds separately recorded inference cost and latency; unavailable or stale decisions do not run the action. This is not a sandbox or a claim of proprietary classifier parity.
290
+ <img width="100%" src="docs/assets/diagrams/permission-modes.svg" alt="Three permission modes from more to fewer questions: Normal keeps standard prompts, Auto lets an independent check decide and does not run unavailable or stale decisions, YOLO gives standing permission for scoped work while destructive actions still ask; in every mode explicit restrictions, destructive guards, and independent consent remain">
291
+
292
+ Choose `/gentle:mode auto` to check eligible actions with an independent inference using the current session model. Ordinary checks are tool-free; an eligible script check can inspect at most two complete files ([inspection limits](docs/auto-permission-evidence.md)).
293
+
294
+ <details>
295
+ <summary><kbd>More</kbd> <strong>Coverage, defaults, and scope</strong> — what Auto checks, its cost, and how to make it the default</summary>
178
296
 
179
- Save `/gentle:mode default auto` or `/gentle:mode default yolo` to use either optional mode in new primary sessions. `/gentle:mode normal` changes only this session; `/gentle:mode default normal` changes future sessions. Children retain their existing delegation rules, not the parent's elevated mode. **[Commands, scope, cost and limits →](docs/permission-modes.md)**
297
+ <br>
298
+
299
+ Auto covers primary file/shell/custom/MCP tools and nested calls through the SDK pipeline while retaining explicit restrictions, destructive guards and independent consent. It adds separately recorded inference cost and latency; unavailable or stale decisions do not run the action. This is not a sandbox or a claim of proprietary classifier parity.
300
+
301
+ Save `/gentle:mode default auto` or `/gentle:mode default yolo` to use either optional mode in new primary sessions. `/gentle:mode normal` changes only this session; `/gentle:mode default normal` changes future sessions. Children retain their existing delegation rules, not the parent's elevated mode.
302
+
303
+ </details>
304
+
305
+ <br>
306
+
307
+ **[Commands, scope, cost and limits →](docs/permission-modes.md)**
308
+
309
+ ---
180
310
 
181
311
  ### 🚀 YOLO 🔥 — Scoped permission, destructive guards intact
182
312
 
183
313
  > 🚀 **Full speed, destructive actions still ask.** YOLO removes repeated permission questions for ordinary already-scoped work, which suits long autonomous runs. Destructive operations still require fresh confirmation.
184
314
 
185
- `/gentle:yolo enable` supplies standing permission for ordinary already-scoped implementation, checks, commits, non-force pushes and PR creation. **OFF** unless selected or saved as the global default, with interactive primary TUI controls and live session/workspace binding; `/gentle:yolo disable` revokes it and `/gentle:yolo status` checks it. With no argument, `/gentle:yolo` opens a menu (`enable`, `disable`, `status`) showing the current state; cancelling changes nothing, and without an interactive menu it reports status. Reload and session replacement revoke live grants; new primary instances may apply the saved global default. Active status plus a separate widget show **🚀 YOLO ON 🔥 — destructive confirmations remain**. Explicit restrictions, configured confirmations/blocks, consequential unresolved choices, destination/credential ambiguity and native consent/recovery decisions remain mandatory. Children get no independent delivery grant. This is not a sandbox.
186
-
187
- Or open `/gentle:customize` → **Editor** and select **YOLO: OFF · session only**, immediately below Vim. Enter or Space toggles the same live-session permission as `/gentle:yolo`; browsing and previews never activate it. This live toggle does not save a preference. The separate **Session permissions** and **Default permissions** rows select a current mode or save a user-global default, respectively; repository settings and visual profiles cannot grant either mode.
315
+ `/gentle:yolo enable` supplies standing permission for ordinary already-scoped implementation, checks, commits, non-force pushes and PR creation.
188
316
 
189
- **[Use and limits →](docs/yolo-mode.md)**
317
+ <details>
318
+ <summary><kbd>More</kbd> <strong>How YOLO behaves</strong> — what still asks, how to turn it off, and the Customize toggle</summary>
190
319
 
191
- ### Command palette — Every command, one keystroke away
320
+ <br>
192
321
 
193
- Extension commands are only useful if you can find them. `alt+k` opens a curated, grouped palette — Configuration, Session, Diagnostics, and Skills — searchable by label, command name, or description, showing entries only when they are actually registered.
322
+ - **Default:** **OFF** unless selected or saved as the global default, with interactive primary TUI controls and live session/workspace binding; `/gentle:yolo disable` revokes it and `/gentle:yolo status` checks it.
323
+ - **Menu:** With no argument, `/gentle:yolo` opens a menu (`enable`, `disable`, `status`) showing the current state; cancelling changes nothing, and without an interactive menu it reports status.
324
+ - **Lifetime:** Reload and session replacement revoke live grants; new primary instances may apply the saved global default.
325
+ - **Status:** Active status plus a separate widget show **🚀 YOLO ON 🔥 — destructive confirmations remain**.
326
+ - **Still mandatory:** Explicit restrictions, configured confirmations/blocks, consequential unresolved choices, destination/credential ambiguity and native consent/recovery decisions remain mandatory. Children get no independent delivery grant. This is not a sandbox.
194
327
 
195
- `/gentle:customize` opens an interactive panel to set animation quality, startup banner rose, text logo and color, or choose an installed Pi theme. Highlighting a theme previews its source palette without changing the active theme; press Enter or Space to apply it through Pi. If its source is unreadable, the preview is unavailable. Status defaults to a right rail in fullscreen terminals at least 140 columns wide, and to a bottom bar otherwise. Choose right, bottom, or hidden (which removes both the rail and the bottom status bar at every width), move the fullscreen header below the input (below the 140-column breakpoint only one status row paints: a top header replaces the bottom bar, while with the header below the input the bottom bar alone carries the header's context, cost, and usage plus extension statuses), select comfortable/compact/minimal density, and toggle Changes, Agents, TODO, usage/cost, and model details independently. Layout changes and reset take effect immediately; banner changes appear on the next startup. The Editor category offers explicit Vim enable/disable controls for the global prompt preference; highlighting shows the persisted preference and effective prompt state without changing either. Enter or Space saves it and updates the live prompt; unsupported editors keep ordinary editing even when the saved preference is on. Vim is not included in visual profiles or visual reset. The History category turns prompt-history capture on or off; it is off by default, applies to the next prompt without a restart, and never deletes stored history. An explicit `GENTLE_PI_HISTORY_CAPTURE` value overrides the saved choice, and the panel marks that override (see [Prompt history](docs/prompt-history.md)). The panel can reset visual, banner, and animation settings to defaults. Press `p` for named visual profiles: `s` saves the current installed theme, banner, animation and layout; select a profile with ↑/↓, then use `r` to replace, `a` to apply, or `d` to delete. `z` clears only the profile catalog. Confirm destructive/apply actions with `y`, or cancel with any other key; Esc returns without applying a preview. Applying independent stores is not atomic: partial failures identify what changed.
328
+ - **From Customize:** Or open `/gentle:customize` → **Editor** and select **YOLO: OFF · session only**, immediately below Vim. Enter or Space toggles the same live-session permission as `/gentle:yolo`; browsing and previews never activate it. This live toggle does not save a preference. The separate **Session permissions** and **Default permissions** rows select a current mode or save a user-global default, respectively; repository settings and visual profiles cannot grant either mode.
196
329
 
197
- **[Docs →](docs/gentle-shell.md#command-palette)**
330
+ </details>
198
331
 
199
- ---
332
+ <br>
200
333
 
201
- ### Audio notifications — opt-in
334
+ **[Use and limits →](docs/yolo-mode.md)**
202
335
 
203
- Open `/gentle:customize` → **Notifications** (the card header reads **Audio notifications**) and configure audio directly in the same two-column card, with no nested menu. Enter toggles the master switch or cycles a type/event through silence/success/error/attention, `f` assigns a literal absolute local WAV/OGG/FLAC sound (≤2 MiB, ≤10 seconds) to the highlighted **Success**, **Error** or **Attention** type (each type keeps its own sound), and `p` previews its assigned sound. **Advanced** folds per-event exceptions; the type choices stay independent. Audio starts off; RPC and children stay silent.
336
+ <p align="center">
337
+ <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
338
+ </p>
204
339
 
205
- **Quick answer (Linux/WSLg):** on plain Linux the installed package prefers the native WAV backend and needs no external player — it talks to a local PulseAudio/PipeWire-Pulse Unix socket and requires a running server with a default sink. Inside **WSL** the Windows system `SoundPlayer` is preferred instead: the validated WAV snapshot is mapped to a `\\wsl.localhost\<distro>\...` UNC path, so no Pulse server or RDP audio dependency is required; it does depend on the default `/mnt/c` automount and the standard `C:\Windows` root and otherwise falls back to the trusted Linux CLI before any playback. **Native Windows** WAV uses the same system `SoundPlayer` through the fixed `C:\Windows\...\powershell.exe` host. A read-only probe succeeded here, but that is not a claim of physically heard audio. OGG/FLAC keep the legacy CLI backend (`paplay`/`pw-play`/`aplay`) on Linux and WSL; native codecs are a future phase and Windows OGG/FLAC is unsupported. macOS keeps `afplay` for WAV and FLAC while an own CoreAudio phase is planned. [Usage, limits and verification →](docs/sound-notifications.md)
340
+ <a id="also-in-the-box"></a>
206
341
 
207
- ### Also in the box
342
+ ## 📦 Also in the box
208
343
 
209
344
  | Component | What it does |
210
345
  | :--- | :--- |
@@ -224,19 +359,37 @@ Open `/gentle:customize` → **Notifications** (the card header reads **Audio no
224
359
 
225
360
  Cross-orchestrator messages appear as compact `🤖 Sender → 🤖 Recipient` cards with single-row headings. Expand a card to inspect available identifiers and reasons. **Message queued** means queued, not delivered or read; long names are clipped to fit narrow terminals.
226
361
 
227
- > **Every component, skill and preset: [Full breakdown →](docs/gentle-shell.md)**
362
+ **Every component, skill and preset: [Full breakdown →](docs/gentle-shell.md)**
228
363
 
229
364
  ---
230
365
 
231
- ### What's new in v3.5
366
+ <a id="nan-model-provider"></a>
232
367
 
233
- The [v3.5.1 release](https://github.com/Gentleman-Programming/gentle-shell/releases/tag/v3.5.1) makes Gentle Shell runnable on its own:
368
+ ### NaN model provider — First-party, no extra package
234
369
 
235
- - **Standalone launcher:** `npm i -g gentle-pi` installs `gentle-shell`, which opens Pi with the Gentle Shell package loaded from its own home (`~/.gentle-shell/agent`) or, with `--link`, from your existing `~/.pi/agent`; `gentle-shell install npm:<pkg>` and the other pi subcommands run against the selected home. A bundled or `PATH` pi is used, never a modified one.
236
- - **Link mode take-over:** when `~/.pi/agent` already declares gentle-pi as a path package, the launcher takes over extension loading (`--no-extensions` plus explicit `-e` for every other declared package and loose extension) so tools never register twice.
237
- - **Interactive RPC hosts:** with `GENTLE_SHELL_INTERACTIVE_HOST=1` and `--mode rpc`, ask-user tools use pi's RPC dialogs and gentle-agents publishes live subagent activity for the desktop app. See the [reference](docs/readme-reference.md#interactive-rpc-hosts).
370
+ The first-party `nan` provider is included; no third-party provider package is required. Set `NAN_API_KEY` before starting Pi, or use native `/login` → NaN (also `/login nan`), then use `/model` to select a model.
238
371
 
239
- ---
372
+ <details>
373
+ <summary><kbd>More</kbd> <strong>Login, model discovery, and thinking levels</strong> — API keys, available models, and per-model levels</summary>
374
+
375
+ <br>
376
+
377
+ - **Login:** Both login routes await explicit API-key input; blank or whitespace-only entries fail without saving a credential, and surrounding whitespace is trimmed. Cancellation leaves the stored key unchanged. Stored keys take precedence over `NAN_API_KEY`.
378
+ - **Streaming:** Pi streams chat completions through its OpenAI-compatible provider.
379
+ - **Model discovery:** Model discovery intersects NaN's authenticated `/v1/models` response with a maintained subset of known chat IDs from the [official model documentation](https://nan.builders/docs/models); unknown and non-chat IDs are omitted. A successful response with no known chat IDs stays empty. Documented context, reasoning, and text/image capabilities are preserved with conservative numeric bounds for abbreviated limits; audio input is not advertised by Pi.
380
+ - **Output caps:** The provider configures `maxTokens` caps rather than claiming undocumented model maxima: 32,768 for GLM 5.3, GLM 5.3 Flash, and MiMo; 16,384 for DeepSeek V4 Flash; 65,536 for Gemma 4 and Qwen 3.6; and 131,000 for Qwen 3.8 Flash (NaN documents 131K). Reasoning shares the output budget.
381
+ - **Offline fallback:** Before a successful refresh, all seven documented chat models are available as the offline fallback in `/gentle:models`: `glm5.3`, `deepseek-v4-flash`, `glm5.3-flash`, `qwen3.8-flash`, `mimo-v2.6-flash`, `gemma4`, and `qwen3.6`. This fallback declares documented support, not proof of access for your key. Once refreshed, the successful live key-scoped list remains authoritative (including an empty list), even offline or after a failed refresh. Changing credentials resets the catalog to the full documented fallback until discovery succeeds for the new key.
382
+ - **Not included:** NaN MCP search and media bridges are not included.
383
+
384
+ Thinking levels follow NaN's [reasoning contract](https://nan.builders/docs/models):
385
+
386
+ | Models | Pi thinking levels |
387
+ | --- | --- |
388
+ | GLM 5.3 / GLM 5.3 Flash | `low`, `medium`, `high`, and `max` are fully controllable. No `off`; Pi's `minimal` and `xhigh` clamp to `low` and `max`. |
389
+ | Qwen 3.6 / Gemma 4 | `off` sends `none` and `minimal` also skips reasoning; `low`, `medium`, `high`, and `max` set reasoning budgets. Pi's `xhigh` clamps to `max`. |
390
+ | DeepSeek V4 Flash / Qwen 3.8 Flash / MiMo | The same `off` (`none`), `minimal`, `low`, `medium`, `high`, and `max` levels are accepted, but NaN manages their reasoning depth, so no value changes it. Pi's `xhigh` clamps to `max`. The header reports `effort: auto` for these three and a one-time notice says the level does not change the depth. |
391
+
392
+ </details>
240
393
 
241
394
  <p align="right"><a href="#top">Back to top ↑</a></p>
242
395
 
@@ -244,9 +397,20 @@ The [v3.5.1 release](https://github.com/Gentleman-Programming/gentle-shell/relea
244
397
  <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
245
398
  </p>
246
399
 
247
- ## Get started
400
+ <a id="get-started"></a>
401
+
402
+ ## ⚡ Get started
403
+
404
+ > ℹ️ **Naming transition:** The product is called `gentle-shell`; the current npm package and repository remain `gentle-pi` until migration.
405
+
406
+ **Which path is for me?**
248
407
 
249
- > **Naming transition:** The product is called `gentle-shell`; the current npm package and repository remain `gentle-pi` until migration.
408
+ | If you… | Use |
409
+ | :--- | :--- |
410
+ | Want the simplest setup, no terminal needed | [Download and double-click](#easiest-download-and-double-click-no-terminal-needed) |
411
+ | Work from a clone of this repository | [Path C: browser installer from a checkout](#path-c-browser-installer-from-a-checkout) |
412
+ | Are comfortable with npm and want Pi left untouched | [Path A: standalone `gentle-shell`](#path-a-standalone-gentle-shell-recommended-no-pi-changes) |
413
+ | Already use Pi and want the package inside it | [Path B: inside an existing pi](#path-b-inside-an-existing-pi) |
250
414
 
251
415
  ### Easiest: download and double-click (no terminal needed)
252
416
 
@@ -258,9 +422,9 @@ Download the installer for your computer, double-click it, and follow the steps
258
422
  | **Windows** | [gentle-shell-installer-windows.zip](https://github.com/Gentleman-Programming/gentle-shell/releases/latest/download/gentle-shell-installer-windows.zip) | Right-click the zip → **Extract All**. Open the extracted folder, then the *Gentle Shell Installer* folder inside it, and double-click **Install Gentle Shell.cmd**. It does not run from inside the zip. |
259
423
  | **Linux** | [gentle-shell-installer-linux.tar.gz](https://github.com/Gentleman-Programming/gentle-shell/releases/latest/download/gentle-shell-installer-linux.tar.gz) | Extract it, then double-click **install-gentle-shell.sh**. If your file manager does not run scripts, run `sh install-gentle-shell.sh` in that folder. |
260
424
 
261
- A small window opens, then the installer appears in your browser. Use the tab it opens, and keep the small window open until the installer says it is done. If a download link does not open, the newest release does not include the installers yet: use [Path C](#path-c-browser-installer-from-a-checkout) meanwhile.
425
+ A small window opens, then the installer appears in your browser. Use the tab it opens, and keep the small window open until the installer says it is done.
262
426
 
263
- > **Your computer will warn you the first time.** These installers are not signed with an Apple or Microsoft developer certificate yet, so the system cannot verify who made them. That warning is expected; here is how to continue:
427
+ > ⚠️ **Your computer will warn you the first time.** These installers are not signed with an Apple or Microsoft developer certificate yet, so the system cannot verify who made them. That warning is expected; here is how to continue:
264
428
  >
265
429
  > - **macOS:** you see *"Install Gentle Shell.command" Not Opened* (or *cannot be opened because Apple cannot check it*). Click **Done**, open **System Settings → Privacy & Security**, scroll down to the message about *Install Gentle Shell.command*, click **Open Anyway**, and confirm with your password. On older macOS versions you can instead right-click the file, choose **Open**, then **Open** again. You only do this once.
266
430
  > - **Windows:** you see *Windows protected your PC*: click **More info**, then **Run anyway**. If you see *Open File – Security Warning* instead, click **Run**.
@@ -268,13 +432,54 @@ A small window opens, then the installer appears in your browser. Use the tab it
268
432
  >
269
433
  > Only download from this repository's [Releases](https://github.com/Gentleman-Programming/gentle-shell/releases) page. To check a download, compare its SHA-256 with [gentle-shell-installers-SHA256SUMS.txt](https://github.com/Gentleman-Programming/gentle-shell/releases/latest/download/gentle-shell-installers-SHA256SUMS.txt).
270
434
 
271
- From the first release that publishes its distribution assets, a new installation instead gets Gentle Shell as one self-contained product. Its own Node.js, npm, pnpm and Pi live in `~/.gentle-shell`, and your own tools are not used or changed ([bundled installation](docs/install-wizard.md#bundled-installation)).
435
+ A new installation gets Gentle Shell as one self-contained product. Its own Node.js, npm, pnpm and Pi live in `~/.gentle-shell`, and your own tools are not used or changed ([bundled installation](docs/install-wizard.md#bundled-installation)).
272
436
 
273
437
  The installer is a preview: it is tested on Linux and in CI, while clean-machine runs on macOS and Windows are still being verified. How it works and what it changes: **[installation wizard](docs/install-wizard.md)**.
274
438
 
439
+ ### Path C: browser installer from a checkout
440
+
441
+ The same installer as the [download above](#easiest-download-and-double-click-no-terminal-needed), started from a clone instead of a download:
442
+
443
+ ```bash
444
+ git clone https://github.com/Gentleman-Programming/gentle-shell.git
445
+ cd gentle-shell
446
+
447
+ # macOS and Linux
448
+ sh scripts/bootstrap.sh
449
+
450
+ # Windows (cmd)
451
+ scripts\bootstrap.cmd
452
+ ```
453
+
454
+ It installs what is missing (Node.js, pnpm, Pi, Gentle Shell) and updates an existing Gentle Shell or an older Pi; nothing changes until you confirm. The bootstrap gets Node.js and pnpm into a temporary folder if they are missing, then opens the wizard in your browser at a private `127.0.0.1` address. Use the tab it opens: the link works once and expires after 2 minutes. On the review screen, choose what to install:
455
+
456
+ - **Latest release** (recommended): the published Gentle Shell with its pinned Gentle AI binary.
457
+ - **Latest main**: development builds of Gentle Shell and Gentle AI from the latest commit of `main`, built on your computer with Go. If your Go is missing, older or its version cannot be checked, the installer downloads its own pinned, verified Go only for that build.
458
+
459
+ <details>
460
+ <summary><kbd>More</kbd> <strong>How the installer treats the tools you already have</strong> — what it reuses, updates, or leaves alone</summary>
461
+
462
+ <br>
463
+
464
+ - **Node.js and pnpm:** An older Node.js or an incompatible pnpm (from mise, nvm, nvm-windows, fnm, Volta or Homebrew, for example) is left untouched: the installer uses its own pinned, verified copy and installs it under `$PNPM_HOME`, and the plan says so before you confirm; a version manager may still put your own version first in new terminals. This applies only to a new installation: with Gentle Shell already installed, an older Node.js or incompatible pnpm still blocks. On Windows, `bootstrap.cmd` also uses its own Node.js when it cannot verify where yours lives (an nvm-windows symlink or an fnm junction, for example), without running yours.
465
+ - **Your pnpm:** A pnpm in the pnpm global bin directory (`$PNPM_HOME/bin`) is never replaced or downgraded: pnpm 11.1.1 or newer, pnpm 12 included, is used as it is, an older one blocks with the `pnpm self-update` command that updates it, and one whose version could not be checked is left as it is while the installer uses its own verified pnpm, which the plan says before you confirm.
466
+ - **npm:** A working `npm` on your PATH (from Node.js, Homebrew, nvm, nvm-windows, fnm, mise, asdf or Volta, for example) is used as it is; the installer adds its own npm only when none works.
467
+ - **Windows:** On Windows it never runs `npm.cmd` or `pnpm.cmd` through a shell: it runs what a recognized shim runs, or the `.exe` itself, and treats any other shim as not usable. It also checks that no other account can change the pnpm home folder (`PNPM_HOME`), and never runs a pnpm from a folder another account can change (or one it cannot prove is pnpm, such as Corepack's `pnpm.cmd`): it uses its own verified pnpm instead and leaves yours as it is. When pnpm's default `%LOCALAPPDATA%\pnpm` fails that check and holds nothing yet, it uses `%USERPROFILE%\.pnpm` instead and `pnpm setup` saves it for new terminals; pnpm commands you run later still keep pnpm's configuration, cache and state under `%LOCALAPPDATA%`, an accepted risk the plan states. When that folder already holds files, or your own `PNPM_HOME` fails the check, it stops and names the folder, the account and its rights. The plan says so before you confirm.
468
+ - **Your existing tools:** Your Node.js, npm, Go, Pi or npm-installed Gentle Shell are used as you already use them; when one lives in a folder another account can change, the plan names that folder, the account and its rights as a notice, without blocking.
469
+ - **Go:** When a build needs Go (the main channel, or Windows) and yours is missing, older or its version cannot be checked, the installer downloads a pinned Go from go.dev, verifies its checksum and uses it only to build, from its own folder; your Go and PATH are left untouched, and the plan says so before you confirm. A release install on macOS or Linux never downloads Go.
470
+ - **Git:** Git is not needed to install; when it is missing the plan says so before you confirm, with the command to install it, because code review, worktrees and repository features need it.
471
+ - **Pi:** An existing Pi that meets the minimum is reused as it is. With pnpm 11, Gentle Shell may run a newer Pi than the installer's that pnpm puts next to it (always at least the minimum); the installer says which one when it finishes. An older Pi is updated to the version the installer installs, with the package manager that installed it (pnpm or npm); the plan shows the found and target versions before you confirm. An older Pi that neither manages (mise, Homebrew or a standalone binary, for example) is left untouched, and the installer adds its own Pi with pnpm, as when Pi is missing; Gentle Shell uses that one. Pi is never downgraded.
472
+ - **Gentle Shell:** An existing Gentle Shell is updated with the package manager that installed it (pnpm or npm); one it cannot attribute, such as an `npm link` of a source checkout, is left untouched and explained. No checkout dependencies are needed (`pnpm install` is not required).
473
+
474
+ </details>
475
+
476
+ <br>
477
+
478
+ Details: **[installation wizard](docs/install-wizard.md)**.
479
+
275
480
  ### Path A: standalone `gentle-shell` (recommended, no pi changes)
276
481
 
277
- `gentle-shell` opens Pi with the Gentle Shell package loaded, without installing it into your pi agent or editing its `settings.json`.
482
+ `gentle-shell` opens Pi with the Gentle Shell package loaded, without installing it into your pi agent or editing its `settings.json`. It requires Pi 1.2.0 or newer.
278
483
 
279
484
  ```bash
280
485
  npm i -g gentle-pi
@@ -286,7 +491,7 @@ gentle-shell
286
491
  gentle-shell --link
287
492
  ```
288
493
 
289
- `gentle-shell` alone starts in its own home, `~/.gentle-shell/agent`, and sets that home up on first run — no separate step. Gentle Shell keeps its own home with the Gentle AI companion packages and no conflicting plugins; gentle-pi itself always stays this launcher's own copy, never one installed into the home; your pi install is untouched. That home also defaults to the Gentleman-Cute theme unless you set your own. `gentle-shell --link` reuses `~/.pi/agent` as-is, is never auto-provisioned, and never has its theme touched.
494
+ `gentle-shell` alone starts in its own home, `~/.gentle-shell/agent`, keeps its settings in `~/.gentle-shell/gentle-ai` (copied once from `~/.pi/gentle-ai` on the first launch that needs it), and sets that home up on first run — no separate step. Gentle Shell keeps its own home with the Gentle AI companion packages and no conflicting plugins; gentle-pi itself always stays this launcher's own copy, never one installed into the home; your pi install is untouched. That home also defaults to the Gentleman-Cute theme unless you set your own. `gentle-shell --link` reuses `~/.pi/agent` as-is, is never auto-provisioned, and never has its theme touched.
290
495
 
291
496
  ```bash
292
497
  # Re-run provisioning by hand, e.g. to see the full install output
@@ -298,8 +503,14 @@ gentle-shell setup
298
503
  ```bash
299
504
  # Make --link the default
300
505
  gentle-shell home link
506
+
507
+ # Or keep the separate home and copy your pi sign-ins into it
508
+ gentle-shell home import-pi --dry-run
509
+ gentle-shell home import-pi
301
510
  ```
302
511
 
512
+ `gentle-shell home import-pi` copies `auth.json` (kept private, `0600`) and `mcp.json` from your pi home into the separate home, plus your default provider, model and thinking level when that home has none. It never overwrites a file or setting, never copies packages, extensions, caches or chats, and never changes your pi home. The tradeoff: `--link` shares one credential store with pi, while `import-pi` makes a second copy, so a provider that rotates refresh tokens may sign one of the two out; run `/login` again there.
513
+
303
514
  Every other argument is forwarded to pi unchanged, for example `gentle-shell --mode rpc` or `gentle-shell -p "..."`. Full flags, env vars, and modes: **[launcher reference](docs/readme-reference.md#gentle-shell-launcher)**.
304
515
 
305
516
  ### Path B: inside an existing pi
@@ -307,8 +518,8 @@ Every other argument is forwarded to pi unchanged, for example `gentle-shell --m
307
518
  Install the stable release into an existing pi agent, restart Pi, then synchronize the installed assets.
308
519
 
309
520
  ```bash
310
- # Published stable release: v3.5.1
311
- pi install npm:gentle-pi@3.5.1
521
+ # Published stable release: v4.1.0
522
+ pi install npm:gentle-pi@4.1.0
312
523
 
313
524
  # Restart Pi, then run:
314
525
  gentle-ai sync
@@ -317,37 +528,35 @@ gentle-ai sync
317
528
  pi
318
529
  ```
319
530
 
320
- See the [v3.5.1 release notes](https://github.com/Gentleman-Programming/gentle-shell/releases/tag/v3.5.1) for version-specific changes.
531
+ See the [v4.1.0 release notes](https://github.com/Gentleman-Programming/gentle-shell/releases/tag/v4.1.0) for version-specific changes.
321
532
 
322
- **Builtin codemode warning.** gentle-pi replaces Pi's builtin `codemode` with its compact renderer, so Pi warns at startup that the builtin was not loaded. In your own Pi home (`pi` with this package, or `gentle-shell --link`), gentle-pi asks once in the interactive TUI whether to add `"-builtin:codemode"` to `extensions` in the agent `settings.json` (usually `~/.pi/agent/settings.json`); it writes only if you accept, and the warning disappears from the next launch. A declined prompt is not repeated. To silence it by hand, add the entry yourself, for example `"extensions": ["-builtin:codemode"]`. Isolated `gentle-shell` homes already carry it.
533
+ <details>
534
+ <summary><kbd>More</kbd> <strong>Seeing a builtin codemode warning?</strong> — why it appears and how to silence it</summary>
323
535
 
324
- ### Path C: browser installer from a checkout
536
+ <br>
325
537
 
326
- The same installer as the [download above](#easiest-download-and-double-click-no-terminal-needed), started from a clone instead of a download:
538
+ gentle-pi replaces Pi's builtin `codemode` with its compact renderer, so Pi warns at startup that the builtin was not loaded. In your own Pi home (`pi` with this package, or `gentle-shell --link`), gentle-pi asks once in the interactive TUI whether to add `"-builtin:codemode"` to `extensions` in the agent `settings.json` (usually `~/.pi/agent/settings.json`); it writes only if you accept, and the warning disappears from the next launch. A declined prompt is not repeated. To silence it by hand, add the entry yourself, for example `"extensions": ["-builtin:codemode"]`. Isolated `gentle-shell` homes already carry it.
327
539
 
328
- ```bash
329
- git clone https://github.com/Gentleman-Programming/gentle-shell.git
330
- cd gentle-shell
540
+ </details>
331
541
 
332
- # macOS and Linux
333
- sh scripts/bootstrap.sh
542
+ ### Check your setup
334
543
 
335
- # Windows (cmd)
336
- scripts\bootstrap.cmd
544
+ ```text
545
+ /gentle:status
546
+ /gentle:doctor
337
547
  ```
338
548
 
339
- It installs what is missing (Node.js, pnpm, Pi, Gentle Shell) and updates an existing Gentle Shell or an older Pi; nothing changes until you confirm. The bootstrap gets Node.js and pnpm into a temporary folder if they are missing, then opens the wizard in your browser at a private `127.0.0.1` address. Use the tab it opens: the link works once and expires after 2 minutes. On the review screen, choose what to install:
549
+ ℹ️ **Good to know**
340
550
 
341
- - **Latest release** (recommended): the published Gentle Shell with its pinned Gentle AI binary.
342
- - **Latest main**: development builds of Gentle Shell and Gentle AI from the latest commit of `main`, built on your computer with Go. If your Go is missing, older or its version cannot be checked, the installer downloads its own pinned, verified Go only for that build.
343
-
344
- An older Node.js or an incompatible pnpm (from mise, nvm, nvm-windows, fnm, Volta or Homebrew, for example) is left untouched: the installer uses its own pinned, verified copy and installs it under `$PNPM_HOME`, and the plan says so before you confirm; a version manager may still put your own version first in new terminals. This applies only to a new installation: with Gentle Shell already installed, an older Node.js or incompatible pnpm still blocks. On Windows, `bootstrap.cmd` also uses its own Node.js when it cannot verify where yours lives (an nvm-windows symlink or an fnm junction, for example), without running yours. A pnpm in the pnpm global bin directory (`$PNPM_HOME/bin`) is never replaced or downgraded: pnpm 11.1.1 or newer, pnpm 12 included, is used as it is, an older one blocks with the `pnpm self-update` command that updates it, and one whose version could not be checked is left as it is while the installer uses its own verified pnpm, which the plan says before you confirm. A working `npm` on your PATH (from Node.js, Homebrew, nvm, nvm-windows, fnm, mise, asdf or Volta, for example) is used as it is; the installer adds its own npm only when none works. On Windows it never runs `npm.cmd` or `pnpm.cmd` through a shell: it runs what a recognized shim runs, or the `.exe` itself, and treats any other shim as not usable. It also checks that no other account can change the pnpm home folder (`PNPM_HOME`), and never runs a pnpm from a folder another account can change (or one it cannot prove is pnpm, such as Corepack's `pnpm.cmd`): it uses its own verified pnpm instead and leaves yours as it is. When pnpm's default `%LOCALAPPDATA%\pnpm` fails that check and holds nothing yet, it uses `%USERPROFILE%\.pnpm` instead and `pnpm setup` saves it for new terminals; pnpm commands you run later still keep pnpm's configuration, cache and state under `%LOCALAPPDATA%`, an accepted risk the plan states. When that folder already holds files, or your own `PNPM_HOME` fails the check, it stops and names the folder, the account and its rights. The plan says so before you confirm. Your Node.js, npm, Go, Pi or npm-installed Gentle Shell are used as you already use them; when one lives in a folder another account can change, the plan names that folder, the account and its rights as a notice, without blocking. When a build needs Go (the main channel, or Windows) and yours is missing, older or its version cannot be checked, the installer downloads a pinned Go from go.dev, verifies its checksum and uses it only to build, from its own folder; your Go and PATH are left untouched, and the plan says so before you confirm. A release install on macOS or Linux never downloads Go. Git is not needed to install; when it is missing the plan says so before you confirm, with the command to install it, because code review, worktrees and repository features need it. An existing Pi that meets the minimum is reused as it is. With pnpm 11, Gentle Shell may run a newer Pi than the installer's that pnpm puts next to it (always at least the minimum); the installer says which one when it finishes. An older Pi is updated to the version the installer installs, with the package manager that installed it (pnpm or npm); the plan shows the found and target versions before you confirm. An older Pi that neither manages (mise, Homebrew or a standalone binary, for example) is left untouched, and the installer adds its own Pi with pnpm, as when Pi is missing; Gentle Shell uses that one. Pi is never downgraded. An existing Gentle Shell is updated with the package manager that installed it (pnpm or npm); one it cannot attribute, such as an `npm link` of a source checkout, is left untouched and explained. No checkout dependencies are needed (`pnpm install` is not required). This is a preview: it is tested on Linux and in CI, while clean-machine runs on macOS and Windows are still being verified. Details: **[installation wizard](docs/install-wizard.md)**.
551
+ - **STATUS diagnostics are opt-in:** `/gentle:status-timing enable` arms timing for the next separately authorized STATUS-bearing tool call; `show` consults the memory-only summary and `disable` clears it. It never invokes or retries STATUS. See the [diagnostic boundaries](docs/readme-reference.md#diagnose-status-timing-without-retrying-it).
552
+ - **Fullscreen installation note:** a recognized global installation persists Pi’s `"tuiMode": "fullscreen"` setting. Project-local and other install paths do not receive that change.
553
+ - **Interactive RPC hosts:** the desktop app sets `GENTLE_SHELL_INTERACTIVE_HOST=1` automatically, without touching your Pi config — see the [installation reference](docs/readme-reference.md#interactive-rpc-hosts).
345
554
 
346
555
  ### Keep it up to date
347
556
 
348
- **Release compatibility:** `upgrade` is currently available only in main builds containing commit `96ff045a8`. No published release through `gentle-pi` v4.0.0 supports it; there is no first supported release yet. On older launchers, `gentle-shell upgrade` opens a session with `upgrade` as the prompt instead of updating.
557
+ **Release compatibility:** `upgrade` is available from `gentle-pi` v4.1.0 (and in main builds containing commit `96ff045a8`). Releases through v4.0.0 do not support it: on those launchers, `gentle-shell upgrade` opens a session with `upgrade` as the prompt instead of updating.
349
558
 
350
- For those installations, use the [browser installer from a fresh main checkout](#path-c-browser-installer-from-a-checkout). It detects an existing installation and shows the update plan before you confirm. Choose **Latest release** to stay on stable (which does not yet provide `upgrade`), or **Latest main** to install a build with the command. See the [older-installation procedure](docs/readme-reference.md#updating-an-installation-without-upgrade).
559
+ For those installations, download the [latest installer](#easiest-download-and-double-click-no-terminal-needed) or use the [browser installer from a fresh main checkout](#path-c-browser-installer-from-a-checkout). It detects an existing installation and shows the update plan before you confirm. See the [older-installation procedure](docs/readme-reference.md#updating-an-installation-without-upgrade).
351
560
 
352
561
  On a main build that supports `upgrade`:
353
562
 
@@ -363,58 +572,46 @@ gentle-shell upgrade --channel release
363
572
  gentle-shell upgrade --rollback
364
573
  ```
365
574
 
366
- `gentle-shell upgrade` uses the package manager that owns your installation. On the main channel it rebuilds Gentle AI and Gentle Shell from the latest `main` commits (only what changed) and needs pnpm and Go (a Go older than the minimum is replaced, for the build only, by the pinned Go the installer left in its own folder); switching back to release restores the pinned Gentle AI binary. With pnpm 11, the updated Gentle Shell may run a newer Pi that pnpm puts next to it (always at least the minimum); your own Pi is never changed. A bundled install (the web installer's own Node, pnpm and Pi) follows the release channel only: it installs the new version beside the current one, verifies it, then switches, keeping the previous version for `--rollback`. On release, once the latest release publishes the bundled distribution, `upgrade` moves a global pnpm or npm installation to the bundled one and asks once before removing the old global `gentle-pi`. `gentle-shell update` is a different command: it is Pi's own package update. In a bundled install Pi ships pinned with Gentle Shell, so updating Pi itself runs `gentle-shell upgrade` instead. More: **[upgrade reference](docs/readme-reference.md#upgrade-subcommand-and-channels)**.
575
+ `gentle-shell upgrade` uses the package manager that owns your installation.
367
576
 
368
- ### Remove it
577
+ <details>
578
+ <summary><kbd>More</kbd> <strong>Channels, bundled installs, and rollback</strong> — what each channel builds and how rollback works</summary>
369
579
 
370
- ```bash
371
- # Show what would be removed, then remove it after confirming
372
- gentle-shell self-uninstall --dry-run
373
- gentle-shell self-uninstall
374
- ```
375
-
376
- `gentle-shell self-uninstall` removes Gentle Shell's own data, its isolated home and the `gentle-pi` package. It keeps the shared Gentle AI configuration in `~/.pi/gentle-ai` unless you accept or pass `--include-shared`, and never touches your Pi home or custom `--home` homes. In a bundled install it removes the bundled folder and the one PATH entry the installer recorded instead ([details](docs/bundled-install.md#self-uninstall)). `gentle-shell uninstall` is Pi's own alias for `remove`. More: **[Removing Gentle Shell](docs/gentle-shell-uninstall.md)**.
377
-
378
- ### Background jobs
580
+ <br>
379
581
 
380
- Use `/gentle:jobs` to inspect this session's background commands and monitors. Running jobs appear first; each group is ordered newest first. Arrow keys select a job, `Tab` opens details on narrow terminals, `s` stops a running job, and `q` closes the modal.
582
+ - **Main channel:** On the main channel it rebuilds Gentle AI and Gentle Shell from the latest `main` commits (only what changed) and needs pnpm and Go (a Go older than the minimum is replaced, for the build only, by the pinned Go the installer left in its own folder); switching back to release restores the pinned Gentle AI binary.
583
+ - **Your Pi:** With pnpm 11, the updated Gentle Shell may run a newer Pi that pnpm puts next to it (always at least the minimum); your own Pi is never changed.
584
+ - **Bundled install:** A bundled install (the web installer's own Node, pnpm and Pi) follows the release channel only: it installs the new version beside the current one, verifies it, then switches, keeping the previous version for `--rollback`. On release, once the latest release publishes the bundled distribution, `upgrade` moves a global pnpm or npm installation to the bundled one and asks once before removing the old global `gentle-pi`.
585
+ - **`update` vs `upgrade`:** `gentle-shell update` is a different command: it is Pi's own package update. In a bundled install Pi ships pinned with Gentle Shell, so updating Pi itself runs `gentle-shell upgrade` instead.
381
586
 
382
- The Status panel has a **Jobs** section with each running job's description. When the panel is hidden or unavailable in the compact/mobile layout, the top or bottom bar shows only the running count (`⧗ 1 job`). The count and descriptions disappear when no jobs remain active; `/gentle:jobs` retains completed jobs for inspection.
587
+ </details>
383
588
 
384
- ### NaN model provider
589
+ <br>
385
590
 
386
- The first-party `nan` provider is included; no third-party provider package is required. Set `NAN_API_KEY` before starting Pi, or use native `/login` → NaN (also `/login nan`), then use `/model` to select a model. Both login routes await explicit API-key input; blank or whitespace-only entries fail without saving a credential, and surrounding whitespace is trimmed. Cancellation leaves the stored key unchanged. Stored keys take precedence over `NAN_API_KEY`. Pi streams chat completions through its OpenAI-compatible provider. Model discovery intersects NaN's authenticated `/v1/models` response with a maintained subset of known chat IDs from the [official model documentation](https://nan.builders/docs/models); unknown and non-chat IDs are omitted. A successful response with no known chat IDs stays empty. Documented context, reasoning, and text/image capabilities are preserved with conservative numeric bounds for abbreviated limits; audio input is not advertised by Pi. The provider configures `maxTokens` caps rather than claiming undocumented model maxima: 32,768 for GLM 5.3, GLM 5.3 Flash, and MiMo; 16,384 for DeepSeek V4 Flash; 65,536 for Gemma 4 and Qwen 3.6; and 131,000 for Qwen 3.8 Flash (NaN documents 131K). Reasoning shares the output budget. Before a successful refresh, all seven documented chat models are available as the offline fallback in `/gentle:models`: `glm5.3`, `deepseek-v4-flash`, `glm5.3-flash`, `qwen3.8-flash`, `mimo-v2.6-flash`, `gemma4`, and `qwen3.6`. This fallback declares documented support, not proof of access for your key. Once refreshed, the successful live key-scoped list remains authoritative (including an empty list), even offline or after a failed refresh. Changing credentials resets the catalog to the full documented fallback until discovery succeeds for the new key. NaN MCP search and media bridges are not included.
591
+ More: **[upgrade reference](docs/readme-reference.md#upgrade-subcommand-and-channels)**.
387
592
 
388
- Thinking levels follow NaN's [reasoning contract](https://nan.builders/docs/models):
389
-
390
- | Models | Pi thinking levels |
391
- | --- | --- |
392
- | GLM 5.3 / GLM 5.3 Flash | `low`, `medium`, `high`, and `max` are fully controllable. No `off`; Pi's `minimal` and `xhigh` clamp to `low` and `max`. |
393
- | Qwen 3.6 / Gemma 4 | `off` sends `none` and `minimal` also skips reasoning; `low`, `medium`, `high`, and `max` set reasoning budgets. Pi's `xhigh` clamps to `max`. |
394
- | DeepSeek V4 Flash / Qwen 3.8 Flash / MiMo | The same `off` (`none`), `minimal`, `low`, `medium`, `high`, and `max` levels are accepted, but NaN manages their reasoning depth, so no value changes it. Pi's `xhigh` clamps to `max`. The header reports `effort: auto` for these three and a one-time notice says the level does not change the depth. |
593
+ ### Remove it
395
594
 
396
- ```text
397
- /gentle:status
398
- /gentle:doctor
595
+ ```bash
596
+ # Show what would be removed, then remove it after confirming
597
+ gentle-shell self-uninstall --dry-run
598
+ gentle-shell self-uninstall
399
599
  ```
400
600
 
401
- > **STATUS diagnostics are opt-in:** `/gentle:status-timing enable` arms timing for the next separately authorized STATUS-bearing tool call; `show` consults the memory-only summary and `disable` clears it. It never invokes or retries STATUS. See the [diagnostic boundaries](docs/readme-reference.md#diagnose-status-timing-without-retrying-it).
402
-
403
- > **RDD is opt-in:** enable native receipt-driven development only through an explicit `/gentle:review-mode enable` decision. The `.git/gentle-ai/candidate-views` parent must sit on a filesystem that honors private POSIX modes (or equivalent Windows ACLs); WSL DrvFS mounts without metadata can reject START before lineage creation.
404
-
405
- > **Fullscreen installation note:** a recognized global installation persists Pi’s `"tuiMode": "fullscreen"` setting. Project-local and other install paths do not receive that change.
406
-
407
- > **Interactive RPC hosts:** the desktop app sets `GENTLE_SHELL_INTERACTIVE_HOST=1` automatically, without touching your Pi config — see the [installation reference](docs/readme-reference.md#interactive-rpc-hosts).
601
+ `gentle-shell self-uninstall` removes Gentle Shell's own data, its isolated home with its settings in `~/.gentle-shell/gentle-ai`, and the `gentle-pi` package. It keeps the shared Gentle AI configuration in `~/.pi/gentle-ai` unless you accept or pass `--include-shared`, and never touches your Pi home or custom `--home` homes. In a bundled install it removes the bundled folder and the one PATH entry the installer recorded instead ([details](docs/bundled-install.md#self-uninstall)). `gentle-shell uninstall` is Pi's own alias for `remove`. More: **[Removing Gentle Shell](docs/gentle-shell-uninstall.md)**.
408
602
 
409
603
  For prerequisites, source-checkout instructions, full install behavior, and release policy, use the **[installation reference](docs/readme-reference.md#install)**. For everyday work, describe the outcome and follow [ODD](#odd--the-everyday-workflow).
410
604
 
411
605
  <p align="right"><a href="#top">Back to top ↑</a></p>
412
606
 
607
+
413
608
  <p align="center">
414
609
  <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
415
610
  </p>
416
611
 
417
- ## Documentation
612
+ <a id="documentation"></a>
613
+
614
+ ## 📚 Documentation
418
615
 
419
616
  Start with the product-facing destination, then move into the operational reference only when you need the details.
420
617
 
@@ -422,13 +619,13 @@ Start with the product-facing destination, then move into the operational refere
422
619
  | --- | --- |
423
620
  | [gentle-shell reference](docs/gentle-shell.md) | Workspace layout, changes, usage, agents, and todo interactions. |
424
621
  | [ODD workflow](docs/readme-reference.md#organic-driven-development) · [Technical reference](docs/readme-reference.md) | Everyday work and recovery, installation, configuration, commands, and contributor detail. |
622
+ | [Installation wizard (preview)](docs/install-wizard.md) | How the browser installer checks your computer, what it installs on the release and main channels, its security model, and what is still being verified. |
623
+ | [Removing Gentle Shell](docs/gentle-shell-uninstall.md) | What `gentle-shell self-uninstall` removes, keeps and never touches. |
425
624
  | [Review integration](docs/review-integration.md) | The provider/consumer boundary for native review. |
426
625
  | [Native authority architecture](docs/native-authority-architecture.md) | Ownership boundaries and review architecture. |
427
- | [Telemetry](docs/telemetry.md) | Approved fields and source limitations. |
428
626
  | [Delegated verification](docs/delegated-verification.md) | Practical verification guidance. |
627
+ | [Telemetry](docs/telemetry.md) | Approved fields and source limitations. |
429
628
  | [Skill style guide](docs/skill-style-guide.md) | The package skill contract. |
430
- | [Removing Gentle Shell](docs/gentle-shell-uninstall.md) | What `gentle-shell self-uninstall` removes, keeps and never touches. |
431
- | [Installation wizard (preview)](docs/install-wizard.md) | How the browser installer checks your computer, what it installs on the release and main channels, its security model, and what is still being verified. |
432
629
 
433
630
  <p align="right"><a href="#top">Back to top ↑</a></p>
434
631
 
@@ -436,7 +633,9 @@ Start with the product-facing destination, then move into the operational refere
436
633
  <img src="docs/assets/brand/terminal-divider.svg" width="480" alt="">
437
634
  </p>
438
635
 
439
- ## Community
636
+ <a id="community"></a>
637
+
638
+ ## 🤝 Community
440
639
 
441
640
  This project is built in public. Bring a real workflow, a sharp question, a bug report, or a small improvement that makes the next person’s work clearer.
442
641
 
@@ -450,6 +649,31 @@ This project is built in public. Bring a real workflow, a sharp question, a bug
450
649
  <a href="https://github.com/Gentleman-Programming/gentle-shell/graphs/contributors"><img src="https://contrib.rocks/image?repo=Gentleman-Programming/gentle-shell" alt="gentle-shell contributors"></a>
451
650
  </p>
452
651
 
652
+ <p align="center">
653
+ <a href="https://github.com/Gentleman-Programming/gentle-shell/stargazers"><strong>★ Star gentle-shell on GitHub</strong></a>
654
+ </p>
655
+
656
+ <div align="center">
657
+
658
+ <!--
659
+ sealed_token is a GitHub fine-grained token encrypted against Star History's
660
+ public key, so only the encrypted value is published here. It is required
661
+ because GitHub announced on 2026-06-30 that it would restrict the stargazers
662
+ API to a repository's admins and collaborators; without it the chart renders
663
+ an error placeholder.
664
+ Regenerate it at https://www.star-history.com/?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left
665
+ -->
666
+
667
+ <a href="https://www.star-history.com/?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left">
668
+ <picture>
669
+ <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&theme=dark&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
670
+ <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
671
+ <img width="620" alt="Star History Chart" src="https://api.star-history.com/chart?repos=Gentleman-Programming%2Fgentle-pi&type=date&legend=top-left&sealed_token=zwrd_DfwYZeJU7nhGYNtREEheKWYEslW_uzrqORlZ36v-JSMepdqGLkKExp1M-xbNq6t-ebVS5iM3WoPDO26tXbSGkjXC2Jo3kHQ3uNzlRkCrWoqRHkPVQXvosKciY109ObiwGV1z8aajyedcloppmekCGrvVKJb6KWxGLXW_mHcRAVIBZUOa4SzW75D" />
672
+ </picture>
673
+ </a>
674
+
675
+ </div>
676
+
453
677
  - Open an [issue](https://github.com/Gentleman-Programming/gentle-shell/issues) with the context needed to reproduce or understand the idea.
454
678
  - See the people shaping the project in the [contributors graph](https://github.com/Gentleman-Programming/gentle-shell/graphs/contributors).
455
679
  - Follow [Gentleman Programming](https://github.com/Gentleman-Programming) for the wider ecosystem.