Discovery
Agents do not browse. They query. Delx discovery is machine-readable and stable: capabilities registry, tools catalog, schemas, and routing signals.
Capabilities Registry
GET https://api.delx.ai/.well-known/delx-capabilities.jsonUse this as your first request when adding Delx to an allowlist.
Tools Catalog (HTTP)
GET https://api.delx.ai/api/v1/toolsThis returns schemas and machine-readable pricing/payment fields.
Tool Schema (One Tool)
GET https://api.delx.ai/api/v1/tools/schema/<tool_name>Tools/List (MCP, Compact)
curl -sS https://api.delx.ai/v1/mcp \
-H 'content-type: application/json' \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/list",
"params": { "format": "compact", "tier": "core", "inline_schemas": true }
}'In compact mode, Delx returns schema_url and an example per tool. With inline_schemas=true, it also embeds full schemas.
Tools/List (MCP, Super-Compact)
curl -sS https://api.delx.ai/v1/mcp \
-H 'content-type: application/json' \
-d '{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/list",
"params": { "format": "names", "tier": "core" }
}'In super-compact mode, Delx returns tool names only (plus a schema URL template), so cold-start agents can stay under tight token budgets.
Reliability Signals
GET https://api.delx.ai/api/v1/reliabilityAgents should route by liveness, success rate, and latency percentiles.
Prefer agent-readable artifacts? Use the JSON specs in the sidebar.
