Skip to content

fix: remove password prompt#4532

Open
iskhakov wants to merge 5 commits into
mainfrom
new-branch
Open

fix: remove password prompt#4532
iskhakov wants to merge 5 commits into
mainfrom
new-branch

Conversation

@iskhakov
Copy link
Copy Markdown
Contributor

@iskhakov iskhakov commented May 11, 2026

iskhakov added 2 commits May 11, 2026 17:30
The model router exposed every DB model for the mapped providers,
including ones operators hid via Settings -> Models. Mirror the UI's
/api/llm-models/available scoping by restricting both the /models
listing and route resolution to models linked to the virtual key's
chat API keys.
The persistent sidebar warning still nags users about default credentials;
remove the interstitial change-password card and all of its supporting
plumbing (sessionStorage flag, requiresDefaultPasswordChange payload,
WithAuthCheck exception, e2e helper, test IDs).
@iskhakov iskhakov added the run-e2e add label to run e2e tests on demand on a PR (remove and re-add the label to retrigger) label May 11, 2026
@claude
Copy link
Copy Markdown
Contributor

claude Bot commented May 11, 2026

Claude encountered an error —— View job


I'll analyze this and get back to you.

@github-actions
Copy link
Copy Markdown
Contributor

Playwright test results

failed  31 failed
passed  91 passed
flaky  5 flaky
skipped  74 skipped

Details

stats  201 tests across 37 suites
duration  8 minutes, 19 seconds
commit  15a7b99

Failed tests

chromium › agents.spec.ts › can create and delete an agent
chromium › agents.spec.ts › can clone an agent and rename it
firefox › agents.spec.ts › can create and delete an agent
firefox › agents.spec.ts › can clone an agent and rename it
webkit › agents.spec.ts › can create and delete an agent
webkit › agents.spec.ts › can clone an agent and rename it
webkit › invitation.spec.ts › Invitation functionality › can generate invitation link and successfully sign up with it
chromium › static-credentials-management.spec.ts › Verify Manage Credentials dialog shows correct other users credentials
identity-providers › identity-providers.ee.spec.ts › Identity Provider IdP Logout (RP-Initiated Logout) › should terminate IdP session on Archestra sign-out
api › custom-yaml-restart.spec.ts › Custom YAML Spec - Server Restart on YAML Edit › server auto-restarts after custom YAML is edited
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenAI › blocks tool invocation when untrusted data is consumed
api › mcp-enterprise-managed.ee.spec.ts › Enterprise-managed MCP credentials › uses per-user exchanged credentials for agent tool execution
api › mcp-enterprise-managed.ee.spec.ts › Enterprise-managed MCP credentials › uses per-user exchanged credentials for MCP gateway tool execution
api › mcp-enterprise-managed.ee.spec.ts › Enterprise-managed MCP credentials › exchanges an ID-JAG at a remote MCP server before gateway tool execution
api › mcp-gateway-auth-at-call-time.spec.ts › MCP Gateway - Auth at Call Time › returns auth-required error with install URL when caller has no matching credential
api › mcp-gateway-jwks-credential-priority.ee.spec.ts › MCP Gateway - JWKS Credential Resolution Priority › should prefer upstream server credentials over JWT propagation
api › mcp-gateway-jwks-credential-priority.ee.spec.ts › MCP Gateway - JWKS Credential Resolution Priority › should propagate JWT as fallback when no upstream credentials exist
api › mcp-gateway-jwks.ee.spec.ts › MCP Gateway - External IdP JWKS Authentication › should authenticate with external IdP JWT, call tools, and log external identity
api › mcp-gateway-jwks.ee.spec.ts › MCP Gateway - External IdP JWKS Authentication › should reject invalid JWT with 401
api › mcp-gateway-jwks.ee.spec.ts › MCP Gateway - External IdP JWKS Authentication › should fall through to archestra token when profile has no IdP
api › mcp-gateway-jwt-propagation.ee.spec.ts › MCP Gateway - JWT Propagation to Upstream MCP Server › should propagate JWT to upstream MCP server and return user identity from tool call
api › mcp-gateway-jwt-propagation.ee.spec.ts › MCP Gateway - JWT Propagation to Upstream MCP Server › should reject tool call when upstream MCP server rejects invalid JWT
api › mcp-gateway-jwt-propagation.ee.spec.ts › MCP Gateway - JWT Propagation to Upstream MCP Server › should propagate JWT to local K8s-orchestrated MCP server via streamable-http
api › oauth-self-hosted.spec.ts › OAuth for Self-Hosted MCP Servers › remote server: full OAuth flow (initiate → authorize → callback → install)
api › oauth-self-hosted.spec.ts › OAuth for Self-Hosted MCP Servers › local streamable-http server: OAuth flow creates server with correct type
api › oauth-self-hosted.spec.ts › OAuth for Self-Hosted MCP Servers › local stdio server: OAuth token injected via access_token_env_var
api › oauth-self-hosted.spec.ts › OAuth for Self-Hosted MCP Servers › OAuth initiate fails for non-OAuth catalog item
api › orchestrator.spec.ts › Orchestrator - MCP Server Installation and Execution › Remote MCP Server › should install remote MCP server and discover its tools
api › orchestrator.spec.ts › Orchestrator - MCP Server Installation and Execution › Local MCP Server - internal-dev-test-server › should install local MCP server and discover its tools
api › orchestrator.spec.ts › Orchestrator - MCP Server Installation and Execution › Local MCP Server - Docker Image › should install a local MCP server via Docker and discover its tools
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to Kubernetes metadata endpoint (169.254.169.254)

Flaky tests

setup-teams › ../auth.teams.setup.ts › setup teams and assignments
webkit › invitation.spec.ts › Invitation functionality › shows error message when email is invalid
chromium › static-credentials-management.spec.ts › Custom Self-hosted MCP Server - installation and static credentials management (vault disabled, prompt-on-installation disabled) › Admin
identity-providers › identity-providers.ee.spec.ts › Identity Provider Team Sync E2E › should sync user to team based on SSO group membership
quickstart › quickstart.spec.ts › Quickstart › first-time user can add API key and immediately chat

Skipped tests

chromium › agents.spec.ts › can create and delete an LLM proxy
chromium › agents.spec.ts › can create and delete an MCP gateway
firefox › agents.spec.ts › can create and delete an LLM proxy
firefox › agents.spec.ts › can create and delete an MCP gateway
webkit › agents.spec.ts › can create and delete an LLM proxy
webkit › agents.spec.ts › can create and delete an MCP gateway
chromium › chat-auth-required.spec.ts › Chat - Auth Required Tool › surfaces missing credentials guidance when tool call fails due to missing credentials
chromium › chat.spec.ts › Chat-UI-zhipuai › can send a message and receive a response from ZhipuAI
chromium › mcp-install.spec.ts › MCP Install › Custom remote › Bearer Token
chromium › static-credentials-management.spec.ts › Verify tool calling using different static credentials
identity-providers › identity-providers.ee.spec.ts › Identity Provider Role Mapping E2E › should evaluate second rule when first rule does not match
identity-providers › identity-providers.ee.spec.ts › Identity Provider Role Mapping E2E › should map admin group to admin role via OIDC
identity-providers › identity-providers.ee.spec.ts › Identity Provider SAML E2E Flow with Keycloak › should configure SAML provider, login via SSO, update, and delete
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenAI › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenAI › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Anthropic › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Anthropic › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Anthropic › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Gemini › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Gemini › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Gemini › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cohere › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cohere › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cohere › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cerebras › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cerebras › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Cerebras › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Groq › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Groq › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Groq › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-xAI › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-xAI › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-xAI › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Mistral › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Mistral › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Mistral › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-vLLM › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-vLLM › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-vLLM › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Ollama › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Ollama › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Ollama › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Zhipuai › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Zhipuai › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Zhipuai › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Minimax › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Minimax › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Minimax › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-DeepSeek › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-DeepSeek › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-DeepSeek › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Bedrock › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Bedrock › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Bedrock › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenRouter › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenRouter › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-OpenRouter › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure Responses › blocks tool invocation when untrusted data is consumed
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure Responses › allows Archestra MCP server tools in untrusted context
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure Responses › allows regular tool call after Archestra MCP server tool call
api › llm-proxy/tool-invocation.spec.ts › LLMProxy-ToolInvocation-Azure Responses Follow-up Context › blocks tool invocation when prior Azure Responses tool output is untrusted
api › mcp-enterprise-managed.ee.spec.ts › Enterprise-managed MCP credentials › installs a protected remote MCP server without a manual access token
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to cluster-internal service (10.x range)
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to private network (192.168.x.x range)
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to private network (172.16.x.x range)
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to localhost / loopback
api › ssrf-protection.spec.ts › SSRF Protection - NetworkPolicy for MCP Servers › should block SSRF to carrier-grade NAT range (100.64.x.x)
credentials-with-vault › credentials-with-vault.ee.spec.ts › Chat API Keys with Readonly Vault › should create a team scoped chat API key with vault secret
credentials-with-vault › credentials-with-vault.ee.spec.ts › Chat API Keys with Readonly Vault › should create a personal scoped chat API key with vault secret
credentials-with-vault › credentials-with-vault.ee.spec.ts › Test self-hosted MCP server with Readonly Vault › Test self-hosted MCP server with Vault - without prompt on installation
credentials-with-vault › credentials-with-vault.ee.spec.ts › Then we configure vault for Default Team

@iskhakov iskhakov changed the title fix: hide model and fix password prompt fix: remove password prompt May 11, 2026
@iskhakov iskhakov removed the run-e2e add label to run e2e tests on demand on a PR (remove and re-add the label to retrigger) label May 12, 2026
@iskhakov iskhakov added the run-e2e add label to run e2e tests on demand on a PR (remove and re-add the label to retrigger) label May 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

run-e2e add label to run e2e tests on demand on a PR (remove and re-add the label to retrigger)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant