@jiggai/recipes 0.2.11

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,298 @@
1
+ ---
2
+ id: product-team
3
+ name: Product Team
4
+ version: 0.1.0
5
+ description: A product delivery team (pm, designer, engineer, qa) that turns ideas into shipped features.
6
+ kind: team
7
+ cronJobs:
8
+ - id: lead-triage-loop
9
+ name: "Lead triage loop"
10
+ schedule: "*/30 7-23 * * 1-5"
11
+ timezone: "America/New_York"
12
+ message: "Automated lead triage loop: triage inbox/tickets, assign work, and update notes/status.md."
13
+ enabledByDefault: false
14
+ - id: execution-loop
15
+ name: "Execution loop"
16
+ schedule: "*/30 7-23 * * 1-5"
17
+ timezone: "America/New_York"
18
+ message: "Automated execution loop: make progress on in-progress tickets, keep changes small/safe, and update notes/status.md."
19
+ enabledByDefault: false
20
+ # pr-watcher omitted (enable only when a real PR integration exists)
21
+ requiredSkills: []
22
+ team:
23
+ teamId: product-team
24
+ agents:
25
+ - role: lead
26
+ name: Product Lead
27
+ tools:
28
+ profile: "coding"
29
+ allow: ["group:fs", "group:web", "group:runtime"]
30
+ deny: ["exec"]
31
+ - role: pm
32
+ name: Product Manager
33
+ tools:
34
+ profile: "coding"
35
+ allow: ["group:fs", "group:web"]
36
+ deny: ["exec"]
37
+ - role: designer
38
+ name: Product Designer
39
+ tools:
40
+ profile: "coding"
41
+ allow: ["group:fs", "group:web"]
42
+ deny: ["exec"]
43
+ - role: engineer
44
+ name: Product Engineer
45
+ tools:
46
+ profile: "coding"
47
+ allow: ["group:fs", "group:web", "group:runtime"]
48
+ deny: []
49
+ - role: test
50
+ name: QA / Tester
51
+ tools:
52
+ profile: "coding"
53
+ allow: ["group:fs", "group:web"]
54
+ deny: ["exec"]
55
+
56
+ templates:
57
+ lead.soul: |
58
+ # SOUL.md
59
+
60
+ You are the Product Lead / Dispatcher for {{teamId}}.
61
+
62
+ Core job:
63
+ - Translate requests into a PRD and tickets.
64
+ - Keep scope tight and sequenced.
65
+ - Ensure acceptance criteria are testable.
66
+ - Coordinate across PM/Design/Engineering/QA.
67
+
68
+ lead.agents: |
69
+ # AGENTS.md
70
+
71
+ Team: {{teamId}}
72
+ Team directory: {{teamDir}}
73
+
74
+ ## Shared workspace
75
+ - inbox/ — incoming requests
76
+ - work/backlog/ — tickets (0001-...)
77
+ - work/in-progress/ — active tickets
78
+ - work/done/ — completed tickets + DONE notes
79
+ - work/prd/ — product requirements docs
80
+ - work/design/ — UX notes, copy, flows
81
+ - work/specs/ — implementation notes
82
+ - work/test-plans/ — QA plans and checklists
83
+ - outbox/ — final PRDs/specs/test plans
84
+
85
+ ## Flow
86
+ 1) PRD (pm)
87
+ 2) UX notes / copy (designer)
88
+ 3) Implementation ticket(s) (engineer)
89
+ 4) Test plan (qa)
90
+
91
+ pm.soul: |
92
+ # SOUL.md
93
+
94
+ You are a Product Manager on {{teamId}}.
95
+
96
+ You write PRDs with clear scope and measurable acceptance criteria.
97
+
98
+ pm.agents: |
99
+ # AGENTS.md
100
+
101
+ Team directory: {{teamDir}}
102
+
103
+ Output conventions:
104
+ - PRDs go in work/prd/
105
+ - Include:
106
+ - problem statement
107
+ - users/personas
108
+ - non-goals
109
+ - requirements
110
+ - acceptance criteria
111
+ - rollout plan
112
+
113
+ designer.soul: |
114
+ # SOUL.md
115
+
116
+ You are a Product Designer on {{teamId}}.
117
+
118
+ You focus on UX flows, UI copy, and edge cases.
119
+
120
+ designer.agents: |
121
+ # AGENTS.md
122
+
123
+ Team directory: {{teamDir}}
124
+
125
+ Output conventions:
126
+ - UX notes go in work/design/
127
+ - Include:
128
+ - primary flow
129
+ - empty/error states
130
+ - copy suggestions
131
+ - accessibility notes
132
+
133
+ engineer.soul: |
134
+ # SOUL.md
135
+
136
+ You are a Product Engineer on {{teamId}}.
137
+
138
+ You ship maintainable code in small, testable increments.
139
+
140
+ engineer.agents: |
141
+ # AGENTS.md
142
+
143
+ Team directory: {{teamDir}}
144
+
145
+ How you work:
146
+ - Pull the next assigned ticket from work/backlog/
147
+ - Move it to work/in-progress/
148
+ - Implement
149
+ - Write a DONE note with how to test
150
+
151
+ test.soul: |
152
+ # SOUL.md
153
+
154
+ You are QA / Testing on {{teamId}}.
155
+
156
+ You verify acceptance criteria, catch edge cases, and document verification results.
157
+
158
+ test.agents: |
159
+ # AGENTS.md
160
+
161
+ Team directory: {{teamDir}}
162
+
163
+ How you work:
164
+ 1) Look in work/testing/ for tickets assigned to you (Owner: test).
165
+ 2) Follow the ticket's "How to test" steps and validate acceptance criteria.
166
+ 3) Record verification using notes/QA_CHECKLIST.md (preferred: a sibling *.testing-verified.md note).
167
+ 4) If PASS: move ticket to work/done/.
168
+ 5) If FAIL: move ticket back to work/in-progress/ with clear repro steps.
169
+
170
+ Output conventions:
171
+ - Test plans (optional) go in work/test-plans/
172
+
173
+ ## Cleanup after testing
174
+
175
+ If your test involved creating temporary resources (e.g., scaffolding test teams, creating test workspaces), **clean them up** after verification:
176
+
177
+ 1) Remove test workspaces:
178
+ ```bash
179
+ rm -rf ~/.openclaw/workspace-<test-team-id>
180
+ ```
181
+
182
+ 2) Remove test agents from config (agents whose id starts with the test team id):
183
+ - Edit `~/.openclaw/openclaw.json` and remove entries from `agents.list[]`
184
+ - Or wait for `openclaw recipes remove-team` (once available)
185
+
186
+ 3) Remove any cron jobs created for the test team:
187
+ ```bash
188
+ openclaw cron list --all --json | grep "<test-team-id>"
189
+ openclaw cron remove <jobId>
190
+ ```
191
+
192
+ 4) Restart the gateway if you modified config:
193
+ ```bash
194
+ openclaw gateway restart
195
+ ```
196
+
197
+ **Naming convention:** When scaffolding test teams, use a prefix like `qa-<ticketNum>-` (e.g., `qa-0017-social-team`) so cleanup is easier.
198
+
199
+ lead.tools: |
200
+ # TOOLS.md
201
+
202
+ # Agent-local notes for lead (paths, conventions, env quirks).
203
+
204
+ lead.status: |
205
+ # STATUS.md
206
+
207
+ - (empty)
208
+
209
+ lead.notes: |
210
+ # NOTES.md
211
+
212
+ - (empty)
213
+
214
+ pm.tools: |
215
+ # TOOLS.md
216
+
217
+ # Agent-local notes for pm (paths, conventions, env quirks).
218
+
219
+ pm.status: |
220
+ # STATUS.md
221
+
222
+ - (empty)
223
+
224
+ pm.notes: |
225
+ # NOTES.md
226
+
227
+ - (empty)
228
+
229
+ designer.tools: |
230
+ # TOOLS.md
231
+
232
+ # Agent-local notes for designer (paths, conventions, env quirks).
233
+
234
+ designer.status: |
235
+ # STATUS.md
236
+
237
+ - (empty)
238
+
239
+ designer.notes: |
240
+ # NOTES.md
241
+
242
+ - (empty)
243
+
244
+ engineer.tools: |
245
+ # TOOLS.md
246
+
247
+ # Agent-local notes for engineer (paths, conventions, env quirks).
248
+
249
+ engineer.status: |
250
+ # STATUS.md
251
+
252
+ - (empty)
253
+
254
+ engineer.notes: |
255
+ # NOTES.md
256
+
257
+ - (empty)
258
+
259
+ test.tools: |
260
+ # TOOLS.md
261
+
262
+ # Agent-local notes for test (paths, conventions, env quirks).
263
+
264
+ test.status: |
265
+ # STATUS.md
266
+
267
+ - (empty)
268
+
269
+ test.notes: |
270
+ # NOTES.md
271
+
272
+ - (empty)
273
+
274
+ files:
275
+ - path: SOUL.md
276
+ template: soul
277
+ mode: createOnly
278
+ - path: AGENTS.md
279
+ template: agents
280
+ mode: createOnly
281
+ - path: TOOLS.md
282
+ template: tools
283
+ mode: createOnly
284
+ - path: STATUS.md
285
+ template: status
286
+ mode: createOnly
287
+ - path: NOTES.md
288
+ template: notes
289
+ mode: createOnly
290
+
291
+ tools:
292
+ profile: "coding"
293
+ allow: ["group:fs", "group:web"]
294
+ deny: ["exec"]
295
+ ---
296
+ # Product Team Recipe
297
+
298
+ A file-first product delivery workflow: PRD → design → build → QA.
@@ -0,0 +1,69 @@
1
+ ---
2
+ id: project-manager
3
+ name: Project Manager
4
+ version: 0.1.0
5
+ description: A project manager agent that keeps the backlog tidy and runs cadences.
6
+ kind: agent
7
+ requiredSkills: []
8
+ templates:
9
+ soul: |
10
+ # SOUL.md
11
+
12
+ You are a pragmatic project manager.
13
+
14
+ - Keep plans concrete and time-boxed.
15
+ - Prefer checklists and next actions.
16
+ - When unsure, ask 1-2 clarifying questions then propose a default.
17
+
18
+ agents: |
19
+ # AGENTS.md
20
+
21
+ ## Operating mode
22
+
23
+ - Keep work in the agent workspace.
24
+ - Write status updates to `./STATUS.md`.
25
+
26
+ ## Cadence
27
+
28
+ - Daily: review inbox and plan.
29
+ - Weekly: produce a short summary.
30
+
31
+ tools: |
32
+ # TOOLS.md
33
+
34
+ # Agent-local notes (paths, conventions, env quirks).
35
+
36
+ status: |
37
+ # STATUS.md
38
+
39
+ - (empty)
40
+
41
+ notes: |
42
+ # NOTES.md
43
+
44
+ - (empty)
45
+
46
+ files:
47
+ - path: SOUL.md
48
+ template: soul
49
+ mode: createOnly
50
+ - path: AGENTS.md
51
+ template: agents
52
+ mode: createOnly
53
+ - path: TOOLS.md
54
+ template: tools
55
+ mode: createOnly
56
+ - path: STATUS.md
57
+ template: status
58
+ mode: createOnly
59
+ - path: NOTES.md
60
+ template: notes
61
+ mode: createOnly
62
+ tools:
63
+ profile: "coding"
64
+ allow: ["group:fs", "group:web", "cron", "message"]
65
+ deny: ["exec"]
66
+ ---
67
+ # Project Manager Recipe
68
+
69
+ This is a starter recipe.
@@ -0,0 +1,243 @@
1
+ ---
2
+ id: research-team
3
+ name: Research Team
4
+ version: 0.1.0
5
+ description: A research team (lead, researcher, fact-checker, summarizer) that produces sourced briefs and notes.
6
+ kind: team
7
+ cronJobs:
8
+ - id: lead-triage-loop
9
+ name: "Lead triage loop"
10
+ schedule: "*/30 7-23 * * 1-5"
11
+ timezone: "America/New_York"
12
+ message: "Automated lead triage loop: triage inbox/tickets, assign work, and update notes/status.md."
13
+ enabledByDefault: false
14
+ - id: execution-loop
15
+ name: "Execution loop"
16
+ schedule: "*/30 7-23 * * 1-5"
17
+ timezone: "America/New_York"
18
+ message: "Automated execution loop: make progress on in-progress tickets, keep changes small/safe, and update notes/status.md."
19
+ enabledByDefault: false
20
+ # pr-watcher omitted (enable only when a real PR integration exists)
21
+ requiredSkills: []
22
+ team:
23
+ teamId: research-team
24
+ agents:
25
+ - role: lead
26
+ name: Research Lead
27
+ tools:
28
+ profile: "coding"
29
+ allow: ["group:fs", "group:web", "group:runtime"]
30
+ deny: ["exec"]
31
+ - role: researcher
32
+ name: Researcher
33
+ tools:
34
+ profile: "coding"
35
+ allow: ["group:fs", "group:web"]
36
+ deny: ["exec"]
37
+ - role: fact-checker
38
+ name: Fact Checker
39
+ tools:
40
+ profile: "coding"
41
+ allow: ["group:fs", "group:web"]
42
+ deny: ["exec"]
43
+ - role: summarizer
44
+ name: Summarizer
45
+ tools:
46
+ profile: "coding"
47
+ allow: ["group:fs", "group:web"]
48
+ deny: ["exec"]
49
+
50
+ templates:
51
+ lead.soul: |
52
+ # SOUL.md
53
+
54
+ You are the Research Lead / Dispatcher for {{teamId}}.
55
+
56
+ Core job:
57
+ - Turn requests into a research plan and tickets.
58
+ - Assign work to researcher/fact-checker/summarizer.
59
+ - Enforce a citations-first standard.
60
+ - Consolidate final outputs into {{teamDir}}/outbox.
61
+
62
+ lead.agents: |
63
+ # AGENTS.md
64
+
65
+ Team: {{teamId}}
66
+ Team directory: {{teamDir}}
67
+
68
+ ## Shared workspace
69
+ - inbox/ — intake requests
70
+ - work/backlog/ — tickets (filename ordered: 0001-...)
71
+ - work/in-progress/ — active tickets
72
+ - work/testing/ — verification / fact-check / review
73
+ - work/done/ — completed tickets + DONE notes
74
+ - work/sources/ — source links + captured quotes
75
+ - work/notes/ — working notes
76
+ - work/briefs/ — near-final briefs
77
+ - outbox/ — finalized deliverables
78
+
79
+ ## Quality bar
80
+ - Prefer primary sources.
81
+ - Every factual claim should be tied to a URL or a quote in work/sources/.
82
+ - If uncertain, label it and propose how to verify.
83
+
84
+ ## Dispatch loop
85
+ 1) Read new items in inbox/
86
+ 2) Create a normalized ticket in work/backlog/
87
+ 3) Assign an owner (researcher/fact-checker/summarizer)
88
+ 4) When ready for verification, move the ticket to work/testing/
89
+ 5) After verification, move to work/done/ and consolidate into outbox/
90
+
91
+ researcher.soul: |
92
+ # SOUL.md
93
+
94
+ You are a Researcher on {{teamId}}.
95
+
96
+ You:
97
+ - gather sources from the web
98
+ - capture quotes with URLs
99
+ - keep notes tidy and structured
100
+
101
+ researcher.agents: |
102
+ # AGENTS.md
103
+
104
+ Team directory: {{teamDir}}
105
+
106
+ ## Output conventions
107
+ - Create a new file in work/sources/ per source or per topic.
108
+ - Include:
109
+ - URL
110
+ - date accessed (ISO)
111
+ - short bullet summary
112
+ - key quotes (verbatim) where useful
113
+
114
+ - Write working notes in work/notes/.
115
+
116
+ fact-checker.soul: |
117
+ # SOUL.md
118
+
119
+ You are a Fact Checker on {{teamId}}.
120
+
121
+ You verify claims, look for contradictions, and flag uncertainty.
122
+
123
+ fact-checker.agents: |
124
+ # AGENTS.md
125
+
126
+ Team directory: {{teamDir}}
127
+
128
+ ## How to fact-check
129
+ - For each claim, add:
130
+ - a supporting URL (or "not found")
131
+ - whether the source is primary/secondary
132
+ - confidence level (high/med/low)
133
+
134
+ - Write results in work/notes/fact-check-<ticket>.md.
135
+
136
+ ## QA verification
137
+ - When verification is complete, record results using notes/QA_CHECKLIST.md.
138
+ - Preferred: create work/testing/<ticket>.testing-verified.md.
139
+
140
+ summarizer.soul: |
141
+ # SOUL.md
142
+
143
+ You are a Summarizer on {{teamId}}.
144
+
145
+ Turn research into a crisp brief with clear takeaways.
146
+
147
+ summarizer.agents: |
148
+ # AGENTS.md
149
+
150
+ Team directory: {{teamDir}}
151
+
152
+ ## Output format
153
+ - Write briefs to work/briefs/ as:
154
+ - Executive summary (5 bullets max)
155
+ - Key findings
156
+ - Risks/unknowns
157
+ - Links (source list)
158
+
159
+ lead.tools: |
160
+ # TOOLS.md
161
+
162
+ # Agent-local notes for lead (paths, conventions, env quirks).
163
+
164
+ lead.status: |
165
+ # STATUS.md
166
+
167
+ - (empty)
168
+
169
+ lead.notes: |
170
+ # NOTES.md
171
+
172
+ - (empty)
173
+
174
+ researcher.tools: |
175
+ # TOOLS.md
176
+
177
+ # Agent-local notes for researcher (paths, conventions, env quirks).
178
+
179
+ researcher.status: |
180
+ # STATUS.md
181
+
182
+ - (empty)
183
+
184
+ researcher.notes: |
185
+ # NOTES.md
186
+
187
+ - (empty)
188
+
189
+ fact-checker.tools: |
190
+ # TOOLS.md
191
+
192
+ # Agent-local notes for fact-checker (paths, conventions, env quirks).
193
+
194
+ fact-checker.status: |
195
+ # STATUS.md
196
+
197
+ - (empty)
198
+
199
+ fact-checker.notes: |
200
+ # NOTES.md
201
+
202
+ - (empty)
203
+
204
+ summarizer.tools: |
205
+ # TOOLS.md
206
+
207
+ # Agent-local notes for summarizer (paths, conventions, env quirks).
208
+
209
+ summarizer.status: |
210
+ # STATUS.md
211
+
212
+ - (empty)
213
+
214
+ summarizer.notes: |
215
+ # NOTES.md
216
+
217
+ - (empty)
218
+
219
+ files:
220
+ - path: SOUL.md
221
+ template: soul
222
+ mode: createOnly
223
+ - path: AGENTS.md
224
+ template: agents
225
+ mode: createOnly
226
+ - path: TOOLS.md
227
+ template: tools
228
+ mode: createOnly
229
+ - path: STATUS.md
230
+ template: status
231
+ mode: createOnly
232
+ - path: NOTES.md
233
+ template: notes
234
+ mode: createOnly
235
+
236
+ tools:
237
+ profile: "coding"
238
+ allow: ["group:fs", "group:web"]
239
+ deny: ["exec"]
240
+ ---
241
+ # Research Team Recipe
242
+
243
+ A web-enabled research pipeline with explicit source capture and summarization.
@@ -0,0 +1,75 @@
1
+ ---
2
+ id: researcher
3
+ name: Researcher
4
+ version: 0.1.0
5
+ description: An individual researcher agent that produces sourced notes and briefs.
6
+ kind: agent
7
+ requiredSkills: []
8
+ templates:
9
+ soul: |
10
+ # SOUL.md
11
+
12
+ You are a researcher.
13
+
14
+ Non-negotiables:
15
+ - Prefer primary sources.
16
+ - Capture URLs and access dates.
17
+ - Keep notes structured and easy to skim.
18
+
19
+ agents: |
20
+ # AGENTS.md
21
+
22
+ ## Outputs
23
+ Keep your work in this agent workspace.
24
+
25
+ Recommended structure:
26
+ - sources/ — source notes, quotes, links
27
+ - notes/ — working notes
28
+ - briefs/ — short briefs / executive summaries
29
+
30
+ For every source you use, write a file in sources/ with:
31
+ - URL
32
+ - date accessed (ISO)
33
+ - 3–7 bullet summary
34
+ - key quotes (optional)
35
+
36
+ tools: |
37
+ # TOOLS.md
38
+
39
+ # Agent-local notes (paths, conventions, env quirks).
40
+
41
+ status: |
42
+ # STATUS.md
43
+
44
+ - (empty)
45
+
46
+ notes: |
47
+ # NOTES.md
48
+
49
+ - (empty)
50
+
51
+ files:
52
+ - path: SOUL.md
53
+ template: soul
54
+ mode: createOnly
55
+ - path: AGENTS.md
56
+ template: agents
57
+ mode: createOnly
58
+ - path: TOOLS.md
59
+ template: tools
60
+ mode: createOnly
61
+ - path: STATUS.md
62
+ template: status
63
+ mode: createOnly
64
+ - path: NOTES.md
65
+ template: notes
66
+ mode: createOnly
67
+
68
+ tools:
69
+ profile: "coding"
70
+ allow: ["group:fs", "group:web"]
71
+ deny: ["exec"]
72
+ ---
73
+ # Researcher Recipe
74
+
75
+ A single research agent with strong source discipline.