/
GO
gotempsh
/
Temps
Search skmtc…
⌘K
Changes
Docs
Sign in
Docs
History
Schema
Filter operations
AdminGate
2
Agents
56
post
/agents/webhook/{webhook_id}
Public webhook endpoint. Authenticated via `X-Webhook-Token` header.
get
/projects/{project_id}/agents
post
/projects/{project_id}/agents
get
/projects/{project_id}/agents/cli-status
get
/projects/{project_id}/agents/runs
get
/projects/{project_id}/agents/runs/latest-for-source
get
/projects/{project_id}/agents/runs/{run_id}
post
/projects/{project_id}/agents/runs/{run_id}/cancel
post
/projects/{project_id}/agents/runs/{run_id}/retry
Retry a completed, failed, cancelled, or no_fix run with the same trigger context. Creates a new run record and spawns the executor.
get
/projects/{project_id}/agents/runs/{run_id}/stream
SSE endpoint for real-time streaming of run events. Polls the agent_run_logs table every 500ms for new entries and streams them. Closes when the run reaches a terminal status.
get
/projects/{project_id}/agents/sandbox-status
post
/projects/{project_id}/agents/smoke-test
Run a smoke test to verify the selected AI CLI works in the environment where agents will actually execute (host or sandbox container). If no `provider_id` is supplied the globally active provider is tested.
get
/projects/{project_id}/agents/{slug}
put
/projects/{project_id}/agents/{slug}
delete
/projects/{project_id}/agents/{slug}
get
/projects/{project_id}/agents/{slug}/runs
post
/projects/{project_id}/agents/{slug}/trigger
get
/projects/{project_id}/autofixer/runs/{run_id}/stream
SSE endpoint: streams run log events in real-time. Polls every 500 ms. Keeps the connection open through "analyzed" and "fix_ready" waiting states; closes only on terminal statuses.
get
/projects/{project_id}/mcp-servers
post
/projects/{project_id}/mcp-servers
get
/projects/{project_id}/mcp-servers/{slug}
put
/projects/{project_id}/mcp-servers/{slug}
delete
/projects/{project_id}/mcp-servers/{slug}
get
/projects/{project_id}/mcp-servers/{slug}/config/{field}
get
/projects/{project_id}/skills
post
/projects/{project_id}/skills
post
/projects/{project_id}/skills/upload
Upload a skill with an archive (tar.gz) — project-scoped.
get
/projects/{project_id}/skills/{slug}
put
/projects/{project_id}/skills/{slug}
delete
/projects/{project_id}/skills/{slug}
get
/projects/{project_id}/skills/{slug}/archive
Download a skill's archive (tar.gz) — project-scoped.
post
/settings/agent-token
Save an encrypted AI provider token for use in sandbox containers.
get
/settings/ai-providers
List the AI provider catalog. Includes per-provider "is a credential configured?" so the settings UI can render configured/not-configured badges without leaking the encrypted credential.
patch
/settings/ai-providers/{provider_id}
Update provider-scoped settings without touching the saved credential. Today that means just `default_model`; future per-provider settings (base URL overrides, request headers, etc.) can land here too without changing the shape of `save_credential`.
post
/settings/ai-providers/{provider_id}/activate
Activate a provider as the platform-wide default. Refuses to activate a provider that doesn't have a credential saved yet — the UI enforces the same rule on the button, but we re-check server-side so a stale tab can't bypass it.
post
/settings/ai-providers/{provider_id}/credential
Save (or replace) a provider's credential. The credential is encrypted with `EncryptionService` and stored inside `agent_sandbox.providers[provider_id].credentials_encrypted`.
post
/settings/ai-providers/{provider_id}/credential/import-local
Import the credential already used by this provider's CLI on the Temps host. Discovery and encryption happen entirely server-side; the plaintext credential is never serialized into the response or browser state.
post
/settings/ai-providers/{provider_id}/credential/verify-saved
Verify a previously imported OpenCode credential against a user-selected model without asking the browser to receive or resubmit the secret.
post
/settings/ai-providers/{provider_id}/models/refresh
Refresh exactly one provider's account-aware model inventory. This is an explicit operation because it starts a CLI process (and, for Claude workspaces, creates or wakes the user's persistent sandbox). Authorization therefore matches the provider's chat execution requirement instead of requiring unrelated settings mutation access. The AI service applies a provider-scoped single-flight and cooldown before doing that work.
post
/settings/ai-providers/{provider_id}/preflight
post
/settings/ai-providers/{provider_id}/smoke
get
/settings/mcp-servers
post
/settings/mcp-servers
get
/settings/mcp-servers/{slug}
put
/settings/mcp-servers/{slug}
delete
/settings/mcp-servers/{slug}
get
/settings/mcp-servers/{slug}/config/{field}
post
/settings/sandbox-rebuild
get
/settings/sandbox-status
get
/settings/skills
post
/settings/skills
post
/settings/skills/upload
Upload a skill with an archive (tar.gz) — global.
get
/settings/skills/{slug}
put
/settings/skills/{slug}
delete
/settings/skills/{slug}
get
/settings/skills/{slug}/archive
Download a skill's archive (tar.gz) — global.
AI Applications
29
AI Chat
43
AI Gateway
14
AI Gateway Admin
12
AI Gateway Pricing
1
AI Gateway Usage
7
AI Provider Status
3
Alarms
12
Alerts
6
Analytics
43
Analytics Ingest Keys
5
API Keys
9
Audit Logs
2
Authentication
29
Autofixer
7
Backups
36
Blob
6
Blob Management
4
Cloud
8
Containers
8
Crons
3
Custom Domains
8
Dashboards
5
Deployment Tokens
6
Deployments
25
DNS
1
DNS Providers
14
DNS Records
4
Domains
19
Email Domains
12
Email Providers
9
Email Tracking
7
Email Validation
1
Emails
5
Environments
2
error-alert-rules
5
error-tracking
10
Events
13
External Images
7
External Services
49
External Services - Query
11
Feature Flags
9
Files
1
Funnels
7
GenAI
2
geo
2
Git Provider Connections
3
Git Providers
35
HTTP Checks
12
Imports
5
Insights
1
Internal DNS
2
IP Access Control
6
KV Management
4
KV Store
7
Load Balancer
5
Logs
11
Metrics
17
Nodes
27
Notification Preferences
4
Notification Providers
15
Notification Routes
7
Observability
2
Operations
1
OTel
15
OTel Facets
4
OTel Ingest
6
Performance
6
Platform
6
Postgres Upgrades
7
Presets
2
Preview Gateway
6
Projects
46
Proxy Logs
14
Public Repositories
6
Repositories
6
Restore
6
Revenue
16
Sandboxes
39
Secrets
7
sentry-compat
6
sentry-ingestor
3
Settings
16
Snapshots
5
source-maps
8
Static Bundles
4
Status Page
32
Teams
13
Telemetry
2
Telemetry Logs
1
Telemetry Metrics
4
Templates
3
Traces
8
Traefik Discovery
6
Traffic Delivery
11
Users
13
Visitor Activity
5
Vulnerability Scans
8
Webhook Deliveries
3
Webhooks
6
Workflows
1
get
/projects/{project_id}/dsns
List all DSNs for a project
post
/projects/{project_id}/dsns
Create a new DSN for a project
post
/projects/{project_id}/dsns/get-or-create
Get or create DSN for a project/environment/deployment combination
post
/projects/{project_id}/dsns/{dsn_id}/regenerate
Regenerate DSN keys (rotate keys)
post
/projects/{project_id}/dsns/{dsn_id}/revoke
Revoke (deactivate) a DSN
Agents
Copy Markdown
Copied
Copy failed
Changes to GET /projects/:project_id/agents/sandbox-status
get
/projects/{project_id}/agents/sandbox-status
Oct 9, 2026
744ab0b80b7b
Revision changes
info
added the required property
microsandbox
to the response with the
200
status