feat: always show the LLM selector; move agent-profile switching to the tools menu

[hieptl]

HUMAN:

End-to-end evidence (beyond unit tests): I ran the full stack locally with an isolated state directory — OH_CANVAS_SAFE_STATE_DIR=<scratch> npm run dev (Vite :3001, uvx agent-server 1.37.0 :18000, ingress :8000) — and drove the UI in Chrome against a fresh onboarding that configured a Claude Code ACP default profile:

• Home pill rendered the constrained Claude model picker (“Available models: Default / Claude Opus 4.8 (1M) / Claude Sonnet 4.6 / Claude Haiku 4.5”) instead of the old agent-profile picker.

• Picking Claude Sonnet 4.6 pre-start persisted into the profile — verified server-side: GET /api/agent-profiles/default → acp_model: "sonnet", revision bump — and the launched conversation ran it (agent replied “hello”; sidebar chip showed Claude Sonnet 4.6).

• “+” menu showed Switch agent profile with both profiles; switching to a codex-test ACP profile flipped the pill to the Codex registry list with GPT‑5.5 as the valid default (while agent_settings still described claude-code — proving the profile-detail sourcing), and switching back restored the profile's remembered Sonnet.

• In the started conversation the “+” menu no longer offered profile switching; the live model switch to Haiku worked and left the profile untouched (acp_model still sonnet server-side).

• Sidebar “+” → blank conversation: pill was the LLM selector, the menu offered Switch agent profile with the “Selecting starts a new conversation” hint, and picking codex-test created and navigated to the replacement conversation (this exposed and led to the mutateAsync navigation fix included here).

• A human has tested these changes.

AGENT:

---

Why

Since Agent Profiles (#1571/#1629), the chat pill becomes an agent-profile picker for every not-yet-started conversation, so users cannot choose an LLM before launching — they must start with the current default and change it afterwards. Making the pre-start LLM picker visible also exposed two latent defects: home ACP state was read from stale agent_settings (activation is pointer-only), and home ACP model picks were persisted to agent_settings_diff.acp_model, which profile-launched conversations never read (agent_profile_id and agent_settings are mutually exclusive launch sources; seeding is one-time).

Summary

• The chat pill is always an LLM selector — provider-constrained ACP model picker in ACP contexts, LLM-profile picker (local) / model chip (cloud) otherwise; the "agent-profile" picker kind is removed and the dead pill component deleted.
• Agent-profile switching moved to the “+” tools menu (Switch agent profile submenu), shown only while starting a new conversation; blank-conversation selection recreates the draft (repo/workspace/plugins carried) and now reliably navigates (mutateAsync — React Query drops mutate-scoped callbacks when the menu unmounts).
• Home ACP provider/model are sourced from the active profile's detail (new useActiveAcpProfileDetail), and pre-start model picks are merge-saved onto the active ACP profile so launches honor them; agent_settings_diff remains only as the no-profiles legacy fallback; cloud members without org-profile permission don't get selectable rows they could only 403 on.

Issue Number

Resolves OpenHands/OpenHands#15435

How to Test

1. npm install, then OH_CANVAS_SAFE_STATE_DIR=$(mktemp -d) npm run dev and open the app on a fresh origin (e.g. http://127.0.0.1:3001).
2. Onboard choosing Claude Code. On home, the pill next to “+” should list only Claude models; pick one, then open Settings → Agent profiles and confirm the profile's model changed.
3. Open the “+” menu → Switch agent profile: create a second ACP profile (e.g. Codex preset) via Manage agent profiles, switch to it, and confirm the pill flips to the GPT model list with that profile's model selected; switch back and confirm the first profile's model is remembered.
4. Send a message: the conversation should run the picked model; the “+” menu must no longer offer Switch agent profile, while the pill still live-switches the model (profile unchanged in Settings).
5. Sidebar “+” → any workspace (blank conversation): pill is an LLM selector; Switch agent profile shows the “Selecting starts a new conversation” hint and picking a profile creates + navigates to a replacement conversation.
6. npm run lint && npx vitest run for the automated suites.

Video/Screenshots

oss-5735.mov

Type

• Bug fix
• Feature
• Refactor
• Breaking change
• Docs / chore

Notes

• Unit status at hand-off: the five touched/new test files pass (37 tests, incl. 13 new: tools-menu gate matrix, useActiveAcpProfileDetail, detail-over-settings sourcing, cloud-member gating, profile persistence + legacy fallback) and tsc --noEmit is clean. The full suite was green (3,960 tests) before the final mutateAsync navigation fix; the post-fix full-suite/lint re-run was deferred to the human tester by request. One known-flaky unrelated test (api-key-entry-screen routing) failed once and passed on re-run.
• Server-side facts this relies on (verified in software-agent-sdk): /switch_acp_model is a 200 persist-only deferral pre-first-run, so blank ACP conversations use the normal live-switch path; stale “409 before first message” comments were corrected.
• Follow-ups (deliberately out of scope, per the issue thread): the named-OpenHands-profile llm_profile_ref layering question (home picks activate the global profile, cosmetic for named profiles), and the cloud-member 403 toast when switching org profiles from the new menu.

---

🐳 Docker images for this PR

• GHCR package: https://github.com/OpenHands/agent-canvas/pkgs/container/agent-canvas

Component	Value
Image	ghcr.io/openhands/agent-canvas
Architectures	amd64, arm64
Agent Server	ghcr.io/openhands/agent-server:1.37.0-python
Automation	openhands-automation==1.3.1
Commit	cd6324a5679161512994a6a2d225a5af7262fa81

Pull (multi-arch manifest)

# Multi-arch manifest — Docker automatically pulls the correct architecture
docker pull ghcr.io/openhands/agent-canvas:sha-cd6324a

Run

docker run -it --rm \
  -p 8000:8000 \
  ghcr.io/openhands/agent-canvas:sha-cd6324a

All tags pushed for this build

ghcr.io/openhands/agent-canvas:sha-cd6324a-amd64
ghcr.io/openhands/agent-canvas:hieptl-oss-5735-amd64
ghcr.io/openhands/agent-canvas:pr-1932-amd64
ghcr.io/openhands/agent-canvas:sha-cd6324a-arm64
ghcr.io/openhands/agent-canvas:hieptl-oss-5735-arm64
ghcr.io/openhands/agent-canvas:pr-1932-arm64
ghcr.io/openhands/agent-canvas:sha-cd6324a
ghcr.io/openhands/agent-canvas:hieptl-oss-5735
ghcr.io/openhands/agent-canvas:pr-1932

About Multi-Architecture Support

• Each tag (e.g., sha-cd6324a) is a multi-arch manifest supporting both amd64 and arm64
• Docker automatically pulls the correct architecture for your platform
• Individual architecture tags (e.g., sha-cd6324a-amd64) are also available if needed

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
agent-canvas	Ready	Preview, Comment	Jul 27, 2026 1:32pm

[jpelletier1]

@hieptl I'd expect profile_1 text in the pill dropdown the name of the LLM profile -- is that the case?

In testing against a Cloud backend, it looks like it's the LLM Model path (openhands/claude-sonnet-5) rather than 'Sonnet', which is the name of my LLM profile.

See examples.

[hieptl]

Hello @jpelletier1,

I've pushed a commit that addresses the issue you reported. You can refer to the video below for additional context.

Screen.Recording.2026-07-27.at.20.18.56.mov

Thank you very much! 🙏

[github-actions]

✅ Mock-LLM E2E Tests

60/60 passed

Commit: cd6324a5 · Workflow run · Test artifacts

Details

Status	Test	Duration
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 1: setup LLM profile and register automation trajectory	6.5s
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 2: create automation and dispatch run via the UI	28.4s
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 3: verify automation and run on the automations page	6.3s
✅	automations/mock-llm-preset-automation.spec.ts › preset automation → slash command conversation › automation card sends the correct slash command to a conversation	15.5s
✅	automations/mock-llm-preset-automation.spec.ts › preset automation → slash command conversation › direct slash command from home page triggers skill activation	13.7s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: fresh install with runtime-injected key › reaches the onboarding modal without pre-seeded localStorage	1.5s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: non-public key rotation › recovers when localStorage has a stale session API key	5.3s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › shows first-run onboarding before the auth screen when no key is configured	1.4s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › rejects an incorrect key with an inline error	1.6s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › allows access after pasting the correct key	1.6s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › skips auth screen for returning user with valid stored key	783ms
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › re-prompts when the server rotates its key (stale localStorage)	1.5s
✅	backends/mock-llm-cross-connect.spec.ts › cross-connect: frontend-only → backend-only › frontend-only connects to a separate backend-only instance	16.9s
✅	backends/mock-llm-cross-connect.spec.ts › cross-connect: frontend-only → multiple backends › connects to two separate backends and switches between them	20.8s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: --frontend-only › serves the frontend but returns 503 for backend routes	7.4s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: --backend-only › serves backend APIs but returns 503 for the frontend root	15.4s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: port conflict › fails with a clear error when the ingress port is occupied	110ms
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: port conflict › starts successfully on a free port after a conflict	6.0s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 1: create an LLM profile pointing at the mock LLM server	6.4s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 2: activate the mock-llm profile and verify settings API	6.3s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 3: run a conversation with the mock LLM	7.3s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 4: resume conversation from sidebar after navigating away	5.7s
✅	conversations/mock-llm-image-upload.spec.ts › mock-LLM image upload › attaching an image embeds it as base64 in the LLM completion call	13.6s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 1: ensure mock LLM profile is configured	6.6s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 2: start conversation and attach workspace metadata	11.7s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 3: git control bar shows workspace pill and git actions	25.3s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 4: files tab can enable diff view for attached workspace	5.8s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 5: browser tab shows empty state	6.4s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 6: files tab defaults to file-tree view without attached workspace	8.3s
✅	home/mock-llm-folder-workspace.spec.ts › mock-LLM folder browser → workspace → conversation › step 1: browse to a folder, add it as a workspace, and launch a conversation with the correct working_dir	8.6s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 1: GitHub card is visible on the MCP marketplace page	5.5s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 2: clicking GitHub add control opens the install modal with correct fields	5.7s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 3: full install flow — fill PAT, submit, verify installed	12.3s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 4: installed GitHub server can be deleted	5.9s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: invalid Slack credentials are blocked with a credential-check error	5.8s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: a valid token missing only a scope still installs (missing_scope is not a credential failure)	5.9s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: an older agent server that omits tool_result still installs (compat)	6.0s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › edit: Test Connection verifies the stored credentials and surfaces a credential failure	5.8s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › edit: Test Connection reports success for valid stored credentials	5.8s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › custom (non-catalog) server: Test Connection attaches no verification probe	5.8s
✅	onboarding/mock-llm-onboarding-happy-path.spec.ts › onboarding happy path › completes the full onboarding flow and launches a conversation	5.0s
✅	onboarding/mock-llm-onboarding-regressions.spec.ts › onboarding recent regressions › keeps the modal open on backdrop click and Escape	1.5s
✅	onboarding/mock-llm-onboarding-regressions.spec.ts › onboarding recent regressions › defaults the LLM setup step to OpenAI GPT-5.5	1.8s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › scopes standalone styles to the agent-server-ui shell	1.4s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › renders critic results on agent messages and finish actions	1.5s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › loads older events when scrolling up	2.0s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › selected workspace persists after navigating away and returning	2.1s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › cleared sessionStorage yields empty workspace selection	1.0s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 1: configure ACP agent via Settings → Agent UI	13.3s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 2: reload and verify ACP settings are persisted in UI	5.7s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 3: start ACP conversation and verify agent reply	6.3s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 4: resume ACP conversation from sidebar after navigating away	5.7s
✅	settings/mock-llm-model-switch.spec.ts › mock-LLM /model slash command › step 1: configure LLM, create switch-target profile, register trajectory	7.2s
✅	settings/mock-llm-model-switch.spec.ts › mock-LLM /model slash command › step 2: start conversation, switch profile via /model, verify switch	7.7s
✅	settings/mock-llm-profile-management.spec.ts › active profile deletion + reconciliation › active profile is deletable and reconciliation activates another profile	8.2s
✅	settings/mock-llm-profile-management.spec.ts › same-model profile identity › chat header shows the correct profile when two profiles share the same model	15.4s
✅	settings/mock-llm-profile-management.spec.ts › OpenHands provider hidden base_url preservation › re-saving an OpenHands profile from Basic view preserves hidden base_url	7.7s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › project skill in workspace/.agents/skills/ triggers on matching keyword	13.7s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › user skill in ~/.openhands/skills/ triggers on matching keyword	8.7s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › deleting a user skill removes it from subsequent conversations	8.7s

_{Posted by the Mock-LLM E2E workflow · results are deterministic (scripted LLM responses)}

[github-actions]

✅ Mock-LLM Docker E2E Test Results

60/60 passed

Commit: cd6324a5 · Workflow run · Test artifacts

Details

Status	Test	Duration
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 1: setup LLM profile and register automation trajectory	8.3s
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 2: create automation and dispatch run via the UI	31.7s
✅	automations/mock-llm-automation.spec.ts › mock-LLM automation lifecycle › step 3: verify automation and run on the automations page	7.1s
✅	automations/mock-llm-preset-automation.spec.ts › preset automation → slash command conversation › automation card sends the correct slash command to a conversation	15.5s
✅	automations/mock-llm-preset-automation.spec.ts › preset automation → slash command conversation › direct slash command from home page triggers skill activation	13.0s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: fresh install with runtime-injected key › reaches the onboarding modal without pre-seeded localStorage	1.5s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: non-public key rotation › recovers when localStorage has a stale session API key	5.3s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › shows first-run onboarding before the auth screen when no key is configured	1.5s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › rejects an incorrect key with an inline error	1.6s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › allows access after pasting the correct key	1.6s
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › skips auth screen for returning user with valid stored key	846ms
✅	backends/mock-llm-auth-modes.spec.ts › auth mode: public gate › re-prompts when the server rotates its key (stale localStorage)	1.4s
✅	backends/mock-llm-cross-connect.spec.ts › cross-connect: frontend-only → backend-only › frontend-only connects to a separate backend-only instance	24.3s
✅	backends/mock-llm-cross-connect.spec.ts › cross-connect: frontend-only → multiple backends › connects to two separate backends and switches between them	21.8s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: --frontend-only › serves the frontend but returns 503 for backend routes	7.4s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: --backend-only › serves backend APIs but returns 503 for the frontend root	15.4s
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: port conflict › fails with a clear error when the ingress port is occupied	105ms
✅	backends/mock-llm-partial-stack.spec.ts › partial stack: port conflict › starts successfully on a free port after a conflict	6.0s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 1: create an LLM profile pointing at the mock LLM server	6.3s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 2: activate the mock-llm profile and verify settings API	8.4s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 3: run a conversation with the mock LLM	6.3s
✅	conversations/mock-llm-conversation.spec.ts › mock-LLM agent-server conversation › step 4: resume conversation from sidebar after navigating away	7.5s
✅	conversations/mock-llm-image-upload.spec.ts › mock-LLM image upload › attaching an image embeds it as base64 in the LLM completion call	15.3s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 1: ensure mock LLM profile is configured	6.6s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 2: start conversation and attach workspace metadata	11.4s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 3: git control bar shows workspace pill and git actions	27.1s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 4: files tab can enable diff view for attached workspace	5.9s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 5: browser tab shows empty state	6.3s
✅	files/mock-llm-files-and-git.spec.ts › files tab, git control bar, and browser tab › step 6: files tab defaults to file-tree view without attached workspace	7.2s
✅	home/mock-llm-folder-workspace.spec.ts › mock-LLM folder browser → workspace → conversation › step 1: browse to a folder, add it as a workspace, and launch a conversation with the correct working_dir	7.4s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 1: GitHub card is visible on the MCP marketplace page	7.7s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 2: clicking GitHub add control opens the install modal with correct fields	5.6s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 3: full install flow — fill PAT, submit, verify installed	12.3s
✅	mcp/mock-llm-mcp-github.spec.ts › MCP GitHub server install flow › step 4: installed GitHub server can be deleted	5.9s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: invalid Slack credentials are blocked with a credential-check error	5.7s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: a valid token missing only a scope still installs (missing_scope is not a credential failure)	5.9s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › install: an older agent server that omits tool_result still installs (compat)	5.9s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › edit: Test Connection verifies the stored credentials and surfaces a credential failure	5.8s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › edit: Test Connection reports success for valid stored credentials	5.8s
✅	mcp/mock-llm-mcp-slack-credentials.spec.ts › MCP Test Connection credential verification (Slack) › custom (non-catalog) server: Test Connection attaches no verification probe	6.0s
✅	onboarding/mock-llm-onboarding-happy-path.spec.ts › onboarding happy path › completes the full onboarding flow and launches a conversation	3.6s
✅	onboarding/mock-llm-onboarding-regressions.spec.ts › onboarding recent regressions › keeps the modal open on backdrop click and Escape	1.5s
✅	onboarding/mock-llm-onboarding-regressions.spec.ts › onboarding recent regressions › defaults the LLM setup step to OpenAI GPT-5.5	1.5s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › scopes standalone styles to the agent-server-ui shell	3.0s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › renders critic results on agent messages and finish actions	1.5s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › loads older events when scrolling up	4.9s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › selected workspace persists after navigating away and returning	2.1s
✅	regressions/mock-llm-ui-regressions.spec.ts › UI regressions › cleared sessionStorage yields empty workspace selection	969ms
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 1: configure ACP agent via Settings → Agent UI	15.4s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 2: reload and verify ACP settings are persisted in UI	5.7s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 3: start ACP conversation and verify agent reply	6.7s
✅	settings/mock-llm-acp-agent.spec.ts › mock-LLM ACP agent conversation › step 4: resume ACP conversation from sidebar after navigating away	5.7s
✅	settings/mock-llm-model-switch.spec.ts › mock-LLM /model slash command › step 1: configure LLM, create switch-target profile, register trajectory	7.3s
✅	settings/mock-llm-model-switch.spec.ts › mock-LLM /model slash command › step 2: start conversation, switch profile via /model, verify switch	6.7s
✅	settings/mock-llm-profile-management.spec.ts › active profile deletion + reconciliation › active profile is deletable and reconciliation activates another profile	8.1s
✅	settings/mock-llm-profile-management.spec.ts › same-model profile identity › chat header shows the correct profile when two profiles share the same model	14.5s
✅	settings/mock-llm-profile-management.spec.ts › OpenHands provider hidden base_url preservation › re-saving an OpenHands profile from Basic view preserves hidden base_url	7.6s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › project skill in workspace/.agents/skills/ triggers on matching keyword	12.7s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › user skill in ~/.openhands/skills/ triggers on matching keyword	7.7s
✅	skills/mock-llm-skills.spec.ts › skill loading: project, user, and deletion › deleting a user skill removes it from subsequent conversations	7.7s

_{Posted by the Mock-LLM E2E workflow · results are deterministic (scripted LLM responses)}

[neubig]

This repository has moved to https://github.com/OpenHands/OpenHands. We’d appreciate it if you re-opened this pull request there. Thank you!