REST API index

hal0-api serves an OpenAI-compatible /v1/* inference surface plus an /api/* management surface that backs the dashboard and the hal0-admin MCP server. This page indexes the /api/* router groups that could be verified directly against source for this pass — for full request/response schemas for a given route, read the OpenAPI spec served live at /api/openapi.json (interactive docs at /api/docs).

Caution

Unlike pre-1.0 docs may claim, auth is not unconditionally removed in this tree. [security].require_auth (overridable via HAL0_REQUIRE_AUTH) and the hal0 auth CLI group (status, rotate, require) are live — see Config schema → [security] and CLI → hal0 auth. Verify the current default (require_auth defaults to null, which resolves to off) before stating a blanket “no auth” or “auth required” claim.

Router groups

Prefix	Purpose
/api/slots	Slot lifecycle — list, create, delete, restart, model swap, resolved-argv provenance, metrics, capacity. See Slot lifecycle.
/api/models	Model registry — list, pull, delete, inspect, scan, catalogue, update checks. See Model roster & benchmarks.
/api/ports	Port allocation view.
/api/stats/hardware	Live hardware probe/stats. See Hardware matrix.
/api/system-info	Host system info.
/api/capabilities	Capability-slot configuration (embed/rerank/stt/tts/vision child slots).
/api/providers	Local provider entries (providers.toml). See Providers, profiles & devices.
/api/upstreams	Remote/cloud upstream entries (upstreams.toml), including credential write and connectivity test.
/api/stacks	Stack bundles — list, create, apply, import/export, snapshot.
/api/profiles	Backend-tuning profiles — list, create, update, import/export.
/api/settings	Root config read/write/reload, JSON schema, apply-plan preview.
/api/settings/models/store	Model store path get/set/migrate.
/api/benchmarks	Benchmark queue, run status, control. See Model roster & benchmarks.
/api/logs	Secret-redacted log tail.
/api/status	Version/build info.
/api/mcp	MCP server lifecycle/introspection — not tool invocation. See MCP tools → MCP server management REST API for the full endpoint table.
/api/comfyui	ComfyUI generation-engine status and image-gen slot control.
/api/install	First-run wizard endpoints, including GET /api/install/curated-models.

Every prefix in this table is backed by at least one route independently confirmed either directly in src/hal0/api/routes/ or via the hal0-admin MCP tool-to-route map (hal0.mcp.admin) — see MCP tools for the exact verb + path for every admin-tool-backed route.

Note

This index was not able to verify the /v1/* OpenAI-compatible inference surface, the “slot as an OpenAI model” addressing scheme, or SSE streaming response shapes in this pass — those need a dedicated read of src/hal0/api/routes/* inference-serving code before publishing exact request/response schemas. Treat any such claims from prior (pre-v1.0) docs as unverified until then.