Agents API

Deploy, configure, and monitor your Hyperfold agents programmatically.

Overview

The Agents API provides full lifecycle management for your commerce agents, including deployment, configuration, scaling, and observability.

EndpointMethodDescription
/v1/agentsGETList all agents
/v1/agents/:idGETGet agent details
/v1/agents/deployPOSTDeploy new agent
/v1/agents/:id/configPATCHUpdate configuration
/v1/agents/:id/logsGETStream logs
/v1/agents/:id/metricsGETGet metrics

List Agents

bash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
# GET /v1/agents
# List all deployed agents
curl -X GET https://api.hyperfold.io/v1/agents \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme"
# Response
{
"data": [
{
"id": "agent_neg_001",
"name": "sales-negotiator",
"type": "negotiator",
"version": "1.2.0",
"status": "running",
"instances": 3,
"created_at": "2025-01-15T10:00:00Z",
"metrics": {
"requests_24h": 4521,
"avg_latency_ms": 340,
"error_rate": 0.003
}
},
{
"id": "agent_ful_001",
"name": "fulfillment-processor",
"type": "fulfillment",
"version": "1.0.0",
"status": "running",
"instances": 2
}
]
}
# GET /v1/agents/:id
# Get agent details
curl -X GET https://api.hyperfold.io/v1/agents/agent_neg_001 \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme"
# Response includes full configuration
{
"id": "agent_neg_001",
"name": "sales-negotiator",
"type": "negotiator",
"version": "1.2.0",
"status": "running",
"config": {
"llm_provider": "openai",
"llm_model": "gpt-4-turbo",
"max_tokens": 4096,
"temperature": 0.7
},
"scaling": {
"min_instances": 2,
"max_instances": 50,
"current_instances": 3,
"target_concurrency": 100
},
"capabilities": ["negotiate", "recommend", "bundle"],
"endpoints": {
"acp": "https://acp.acme.hyperfold.app/negotiator"
}
}

Deploy Agent

bash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
# POST /v1/agents/deploy
# Deploy a new agent or update existing
curl -X POST https://api.hyperfold.io/v1/agents/deploy \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme" \
-H "Content-Type: application/json" \
-d '{
"name": "sales-negotiator",
"type": "negotiator",
"source": {
"type": "git",
"repository": "https://github.com/acme/agents",
"branch": "main",
"path": "negotiator/"
},
"config": {
"llm_provider": "openai",
"llm_model": "gpt-4-turbo",
"temperature": 0.7,
"system_prompt_override": null
},
"scaling": {
"min_instances": 2,
"max_instances": 50,
"target_concurrency": 100
},
"environment": {
"CUSTOM_VAR": "value"
}
}'
# Response (202 Accepted)
{
"deployment_id": "deploy_xyz789",
"agent_id": "agent_neg_001",
"status": "deploying",
"progress": {
"stage": "building",
"percent": 25
},
"estimated_completion": "2025-01-20T10:05:00Z"
}
# GET /v1/agents/deploy/:deployment_id
# Check deployment status
curl -X GET https://api.hyperfold.io/v1/agents/deploy/deploy_xyz789 \
-H "Authorization: Bearer hf_live_xxx"
# Response
{
"deployment_id": "deploy_xyz789",
"agent_id": "agent_neg_001",
"status": "completed",
"version": "1.2.1",
"completed_at": "2025-01-20T10:04:30Z",
"logs_url": "https://console.hyperfold.io/deployments/deploy_xyz789/logs"
}

Deployment Sources

Source TypeDescription
gitDeploy from Git repository
dockerDeploy from container image
templateUse built-in agent template
uploadUpload source directly

Configure

bash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
# PATCH /v1/agents/:id/config
# Update agent configuration (hot reload)
curl -X PATCH https://api.hyperfold.io/v1/agents/agent_neg_001/config \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme" \
-H "Content-Type: application/json" \
-d '{
"llm_model": "gpt-4o",
"temperature": 0.5,
"max_discount_percent": 25,
"negotiation_style": "friendly"
}'
# Response
{
"agent_id": "agent_neg_001",
"config_version": 3,
"changes": [
{"field": "llm_model", "from": "gpt-4-turbo", "to": "gpt-4o"},
{"field": "temperature", "from": 0.7, "to": 0.5}
],
"applied_at": "2025-01-20T10:10:00Z",
"rollout": "immediate"
}
# PATCH /v1/agents/:id/scaling
# Update scaling configuration
curl -X PATCH https://api.hyperfold.io/v1/agents/agent_neg_001/scaling \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme" \
-H "Content-Type: application/json" \
-d '{
"min_instances": 5,
"max_instances": 100,
"target_concurrency": 150
}'
# POST /v1/agents/:id/restart
# Restart all agent instances
curl -X POST https://api.hyperfold.io/v1/agents/agent_neg_001/restart \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme" \
-d '{"strategy": "rolling"}'
Configuration changes are applied immediately to running instances without downtime. Use the restart endpoint for changes requiring a full reload.

Logs & Metrics

bash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
# GET /v1/agents/:id/logs
# Stream agent logs
curl -X GET "https://api.hyperfold.io/v1/agents/agent_neg_001/logs?\
since=1h&\
level=info&\
limit=100" \
-H "Authorization: Bearer hf_live_xxx" \
-H "X-Project-ID: proj_acme"
# Response
{
"logs": [
{
"timestamp": "2025-01-20T10:15:30.123Z",
"level": "info",
"message": "Session started",
"context": {
"session_id": "sess_abc123",
"customer_tier": "gold"
}
},
{
"timestamp": "2025-01-20T10:15:30.456Z",
"level": "info",
"message": "Price quote generated",
"context": {
"product_id": "prod_xyz",
"offered": 150.00,
"counter": 162.00
}
}
],
"meta": {
"count": 100,
"has_more": true,
"next_cursor": "cursor_xyz"
}
}
# GET /v1/agents/:id/metrics
# Get agent performance metrics
curl -X GET "https://api.hyperfold.io/v1/agents/agent_neg_001/metrics?\
period=24h&\
granularity=1h" \
-H "Authorization: Bearer hf_live_xxx"
# Response
{
"agent_id": "agent_neg_001",
"period": "24h",
"summary": {
"total_requests": 4521,
"success_rate": 0.997,
"avg_latency_ms": 340,
"p95_latency_ms": 890,
"conversions": 1247,
"revenue": 187450.00
},
"timeseries": [
{"timestamp": "2025-01-20T00:00:00Z", "requests": 120, "latency_ms": 320},
{"timestamp": "2025-01-20T01:00:00Z", "requests": 85, "latency_ms": 310}
]
}

Log Levels

LevelDescription
debugDetailed debugging information
infoGeneral operational events
warnWarning conditions
errorError events
Next: Manage payments with the Finance API.