data-machine

Name: data-machine
Author: Sarai-Chinwag

64stars7forks

How to install data-machine?

sarai-chinwag/wp-openclaw/data-machine

$npx -y skills add Sarai-Chinwag/wp-openclaw --skill data-machine

Installs into the current project.

Summary

Self-scheduling execution layer for autonomous task orchestration. Use for queuing tasks, chaining pipeline executions, scheduling recurring work, and 24/7 autonomous operation via Agent Ping webhooks.

Files · 1

View on GitHub

SKILL.md

1	# Data Machine Skill
2
3	A self-scheduling execution layer for AI agents. Not just content automation — it's how agents schedule themselves to achieve goals autonomously.
4
5	## When to Use This Skill
6
7	Use this skill when:
8	- Setting up automated workflows (content generation, publishing, notifications)
9	- Creating self-scheduling patterns (reminders, recurring tasks)
10	- Building multi-phase projects with queued task progression
11	- Configuring Agent Ping webhooks to trigger external agents
12
13	---
14
15	## Core Philosophy
16
17	Data Machine is designed with AI agents as primary users. It functions as a reminder system + task manager + workflow executor all in one.
18
19	### Three Key Concepts
20
21	1. Flows operate on schedules — Configure "ping me at X time to do Y"
22	2. Step-level prompt queues — Each ping can be a different task instruction
23	3. Multiple purpose-specific flows — Separate flows for separate concerns
24
25	### Mental Model
26
27	\| Role \| How It Works \|
28	\|------\|--------------\|
29	\| Reminder System \| Flows run on schedules (daily, hourly, cron) and ping the agent \|
30	\| Task Manager \| Queues hold task backlog; each run pops the next task \|
31	\| Workflow Executor \| Pipeline steps execute work (AI generation, publishing, API calls) \|
32
33	---
34
35	## Architecture Overview
36
37	### Execution Model
38
39	```
40	Pipeline (template) → Flow (instance) → Job (execution)
41	```
42
43	- Pipeline: Reusable workflow template with steps
44	- Flow: Instance of a pipeline with specific configuration and schedule
45	- Job: Single execution of a flow
46
47	### Step Types
48
49	\| Type \| Purpose \| Has Queue \|
50	\|------\|---------\|-----------\|
51	\| `fetch` \| Import data (RSS, Sheets, Files, Reddit) \| No \|
52	\| `ai` \| Process with AI (multi-turn, tools) \| Yes \|
53	\| `publish` \| Output (WordPress, Twitter, Discord) \| No \|
54	\| `update` \| Modify existing content \| No \|
55	\| `agent_ping` \| Webhook to external agents \| Yes \|
56
57	### Scheduling Options
58
59	Configure via `scheduling_config` in the flow:
60
61	\| Interval \| Behavior \|
62	\|----------\|----------\|
63	\| `manual` \| Only runs when triggered via UI or CLI \|
64	\| `daily` \| Runs once per day \|
65	\| `hourly` \| Runs once per hour \|
66	\| `{"cron": "0 9 * * 1"}` \| Cron expression (e.g., Mondays at 9am) \|
67
68	---
69
70	## Working with Flows
71
72	### List Flows
73
74	```bash
75	wp datamachine flows list
76	```
77
78	### Get Flow Details
79
80	```bash
81	wp datamachine flows get <flow_id>
82	```
83
84	### Run a Flow Manually
85
86	```bash
87	wp datamachine flows run <flow_id>
88	```
89
90	### Check Job Status
91
92	```bash
93	wp datamachine jobs list --limit=10
94	```
95
96	---
97
98	## Prompt Queues
99
100	Both AI and Agent Ping steps support queues via `QueueableTrait`. If the configured prompt is empty and `queue_enabled` is true, the step pops from its queue.
101
102	This enables varied task instructions per execution — not the same prompt every time.
103
104	### Queue Management
105
106	```bash
107	# Add to queue
108	wp datamachine flows queue add <flow_id> "Task instruction here"
109
110	# List queue contents
111	wp datamachine flows queue list <flow_id>
112
113	# Clear queue
114	wp datamachine flows queue clear <flow_id>
115	```
116
117	### Chaining Pattern
118
119	When an agent receives a ping, it should:
120	1. Execute the immediate task
121	2. Queue the next logical task (if continuation needed)
122	3. Let the cycle continue
123
124	```
125	Ping: "Phase 1: Design the architecture"
126	→ Agent designs, writes DESIGN.md
127	→ Agent queues: "Phase 2: Implement schema per DESIGN.md"
128
129	Ping: "Phase 2: Implement schema per DESIGN.md"
130	→ Agent implements
131	→ Agent queues: "Phase 3: Build API endpoints"
132	```
133
134	The queue becomes the agent's persistent project memory — multi-phase work is tracked in the queue, not held in context.
135
136	---
137
138	## Purpose-Specific Flows
139
140	Critical pattern: Don't try to do everything in one flow. Create separate flows for separate concerns:
141
142	```
143	Flow: Content Generation (queue-driven)
144	→ AI Step (pops topic from queue) → Publish → Agent Ping
145
146	Flow: Content Ideation (daily)
147	→ Agent Ping: "Review analytics, add topics to content queue"
148
149	Flow: Weekly Review (cron: Monday 9am)
150	→ Agent Ping: "Analyze last week's performance"
151
152	Flow: Coding Tasks (manual, queue-driven)
153	→ Agent Ping (pops from queue): specific coding task instructions
154	```
155
156	Each flow has its own:
157	- Schedule: When it runs
158	- Queue: Task backlog specific to that workflow
159	- Purpose: Single responsibility, clear scope
160
161	---
162
163	## Agent Ping Configuration
164
165	Agent Ping steps send webhooks to external agent frameworks (OpenClaw, LangChain, custom handlers).
166
167	### Handler Configuration
168
169	- `webhook_url`: Where to send the ping
170	- `prompt`: Static prompt, or leave empty to use queue
171	- `queue_enabled`: Whether to pop from queue when prompt is empty
172
173	### Webhook Payload
174
175	The ping includes:
176	- Flow and job co

People who liked this also viewed

6 picks

Type

data-machine

64stars7forks

View on GitHub

How to install data-machine?

sarai-chinwag/wp-openclaw/data-machine

$npx -y skills add Sarai-Chinwag/wp-openclaw --skill data-machine

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Data Machine Skill
2
3	A self-scheduling execution layer for AI agents. Not just content automation — it's how agents schedule themselves to achieve goals autonomously.
4
5	## When to Use This Skill
6
7	Use this skill when:
8	- Setting up automated workflows (content generation, publishing, notifications)
9	- Creating self-scheduling patterns (reminders, recurring tasks)
10	- Building multi-phase projects with queued task progression
11	- Configuring Agent Ping webhooks to trigger external agents
12
13	---
14
15	## Core Philosophy
16
17	Data Machine is designed with AI agents as primary users. It functions as a reminder system + task manager + workflow executor all in one.
18
19	### Three Key Concepts
20
21	1. Flows operate on schedules — Configure "ping me at X time to do Y"
22	2. Step-level prompt queues — Each ping can be a different task instruction
23	3. Multiple purpose-specific flows — Separate flows for separate concerns
24
25	### Mental Model
26
27	\| Role \| How It Works \|
28	\|------\|--------------\|
29	\| Reminder System \| Flows run on schedules (daily, hourly, cron) and ping the agent \|
30	\| Task Manager \| Queues hold task backlog; each run pops the next task \|
31	\| Workflow Executor \| Pipeline steps execute work (AI generation, publishing, API calls) \|
32
33	---
34
35	## Architecture Overview
36
37	### Execution Model
38
39	```
40	Pipeline (template) → Flow (instance) → Job (execution)
41	```
42
43	- Pipeline: Reusable workflow template with steps
44	- Flow: Instance of a pipeline with specific configuration and schedule
45	- Job: Single execution of a flow
46
47	### Step Types
48
49	\| Type \| Purpose \| Has Queue \|
50	\|------\|---------\|-----------\|
51	\| `fetch` \| Import data (RSS, Sheets, Files, Reddit) \| No \|
52	\| `ai` \| Process with AI (multi-turn, tools) \| Yes \|
53	\| `publish` \| Output (WordPress, Twitter, Discord) \| No \|
54	\| `update` \| Modify existing content \| No \|
55	\| `agent_ping` \| Webhook to external agents \| Yes \|
56
57	### Scheduling Options
58
59	Configure via `scheduling_config` in the flow:
60
61	\| Interval \| Behavior \|
62	\|----------\|----------\|
63	\| `manual` \| Only runs when triggered via UI or CLI \|
64	\| `daily` \| Runs once per day \|
65	\| `hourly` \| Runs once per hour \|
66	\| `{"cron": "0 9 * * 1"}` \| Cron expression (e.g., Mondays at 9am) \|
67
68	---
69
70	## Working with Flows
71
72	### List Flows
73
74	```bash
75	wp datamachine flows list
76	```
77
78	### Get Flow Details
79
80	```bash
81	wp datamachine flows get <flow_id>
82	```
83
84	### Run a Flow Manually
85
86	```bash
87	wp datamachine flows run <flow_id>
88	```
89
90	### Check Job Status
91
92	```bash
93	wp datamachine jobs list --limit=10
94	```
95
96	---
97
98	## Prompt Queues
99
100	Both AI and Agent Ping steps support queues via `QueueableTrait`. If the configured prompt is empty and `queue_enabled` is true, the step pops from its queue.
101
102	This enables varied task instructions per execution — not the same prompt every time.
103
104	### Queue Management
105
106	```bash
107	# Add to queue
108	wp datamachine flows queue add <flow_id> "Task instruction here"
109
110	# List queue contents
111	wp datamachine flows queue list <flow_id>
112
113	# Clear queue
114	wp datamachine flows queue clear <flow_id>
115	```
116
117	### Chaining Pattern
118
119	When an agent receives a ping, it should:
120	1. Execute the immediate task
121	2. Queue the next logical task (if continuation needed)
122	3. Let the cycle continue
123
124	```
125	Ping: "Phase 1: Design the architecture"
126	→ Agent designs, writes DESIGN.md
127	→ Agent queues: "Phase 2: Implement schema per DESIGN.md"
128
129	Ping: "Phase 2: Implement schema per DESIGN.md"
130	→ Agent implements
131	→ Agent queues: "Phase 3: Build API endpoints"
132	```
133
134	The queue becomes the agent's persistent project memory — multi-phase work is tracked in the queue, not held in context.
135
136	---
137
138	## Purpose-Specific Flows
139
140	Critical pattern: Don't try to do everything in one flow. Create separate flows for separate concerns:
141
142	```
143	Flow: Content Generation (queue-driven)
144	→ AI Step (pops topic from queue) → Publish → Agent Ping
145
146	Flow: Content Ideation (daily)
147	→ Agent Ping: "Review analytics, add topics to content queue"
148
149	Flow: Weekly Review (cron: Monday 9am)
150	→ Agent Ping: "Analyze last week's performance"
151
152	Flow: Coding Tasks (manual, queue-driven)
153	→ Agent Ping (pops from queue): specific coding task instructions
154	```
155
156	Each flow has its own:
157	- Schedule: When it runs
158	- Queue: Task backlog specific to that workflow
159	- Purpose: Single responsibility, clear scope
160
161	---
162
163	## Agent Ping Configuration
164
165	Agent Ping steps send webhooks to external agent frameworks (OpenClaw, LangChain, custom handlers).
166
167	### Handler Configuration
168
169	- `webhook_url`: Where to send the ping
170	- `prompt`: Static prompt, or leave empty to use queue
171	- `queue_enabled`: Whether to pop from queue when prompt is empty
172
173	### Webhook Payload
174
175	The ping includes:
176	- Flow and job co

People who liked this also viewed

6 picks

Type