dogfood

Name: dogfood
Author: callstackincubator

1.6kstars111forks

How to install dogfood?

callstackincubator/agent-skills/dogfood

$npx -y skills add callstackincubator/agent-skills --skill dogfood

Installs into the current project.

Summary

Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to "dogfood", "QA", "exploratory test", "find issues", "bug hunt", or "test this app" on mobile. Produces a structured report with reproducibl

Files · 1

View on GitHub

SKILL.md

1	# Dogfood (agent-device)
2
3	Systematically explore a mobile app, find issues, and produce a report with full reproduction evidence for every finding.
4
5	## Setup
6
7	Only the Target app is required. Everything else has sensible defaults.
8
9	\| Parameter \| Default \| Example override \|
10	\| -------------------- \| ----------------------------------------------------------- \| -------------------------------------------- \|
11	\| Target app \| _(required)_ \| `Settings`, `com.example.app`, deep link URL \|
12	\| Platform \| Infer from user context; otherwise ask (`ios` or `android`) \| `--platform ios` \|
13	\| Session name \| Slugified app/platform (for example `settings-ios`) \| `--session my-session` \|
14	\| Output directory \| `./dogfood-output/` \| `Output directory: /tmp/mobile-qa` \|
15	\| Scope \| Full app \| `Focus on onboarding and profile` \|
16	\| Authentication \| None \| `Sign in to user@example.com` \|
17
18	If the user gives enough context to start, begin immediately with defaults. Ask follow-up only when a required detail is missing (for example platform or credentials).
19
20	Prefer direct `agent-device` binary when available.
21
22	## Workflow
23
24	```
25	1. Initialize Set up session, output dirs, report file
26	2. Launch/Auth Open app and sign in if needed
27	3. Orient Capture initial snapshot and map navigation
28	4. Explore Systematically test flows and states
29	5. Document Record reproducible evidence per issue
30	6. Wrap up Reconcile summary, close session
31	```
32
33	### 1. Initialize
34
35	```bash
36	mkdir -p {OUTPUT_DIR}/screenshots {OUTPUT_DIR}/videos
37	cp {SKILL_DIR}/templates/dogfood-report-template.md {OUTPUT_DIR}/report.md
38	```
39
40	### 2. Launch/Auth
41
42	Start a named session and launch target app:
43
44	```bash
45	agent-device --session {SESSION} open {TARGET_APP} --platform {PLATFORM}
46	agent-device --session {SESSION} snapshot -i
47	```
48
49	If login is required:
50
51	```bash
52	agent-device --session {SESSION} snapshot -i
53	agent-device --session {SESSION} fill @e1 "{EMAIL}"
54	agent-device --session {SESSION} fill @e2 "{PASSWORD}"
55	agent-device --session {SESSION} press @e3
56	agent-device --session {SESSION} wait 1000
57	agent-device --session {SESSION} snapshot -i
58	```
59
60	For OTP/email codes: ask the user, wait for input, then continue.
61
62	### 3. Orient
63
64	Capture initial evidence and navigation anchors:
65
66	```bash
67	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/initial.png
68	agent-device --session {SESSION} snapshot -i
69	```
70
71	Map top-level navigation, tabs, and key workflows before deep testing.
72
73	### 4. Explore
74
75	Read [references/issue-taxonomy.md](references/issue-taxonomy.md) for severity/category calibration.
76
77	Strategy:
78
79	- Move through each major app area (tabs, drawers, settings pages).
80	- Test core journeys end-to-end (create, edit, delete, submit, recover).
81	- Validate edge states (empty/error/loading/offline/permissions denied).
82	- Use `diff snapshot -i` after UI transitions to avoid stale refs.
83	- Periodically capture `logs path` and inspect the app log when behavior looks suspicious.
84
85	Useful commands per screen:
86
87	```bash
88	agent-device --session {SESSION} snapshot -i
89	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/{screen-name}.png
90	agent-device --session {SESSION} appstate
91	agent-device --session {SESSION} logs path
92	```
93
94	### 5. Document Issues (Repro-First)
95
96	Explore and document in one pass. When you find an issue, stop and fully capture evidence before continuing.
97
98	#### Interactive/behavioral issues
99
100	Use video + step screenshots:
101
102	1. Start recording:
103
104	```bash
105	agent-device --session {SESSION} record start {OUTPUT_DIR}/videos/issue-{NNN}-repro.mp4
106	```
107
108	2. Reproduce with visible pacing. Capture each step:
109
110	```bash
111	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-1.png
112	sleep 1
113	# perform action
114	sleep 1
115	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-2.png
116	```
117
118	3. Capture final broken state:
119
120	```bash
121	sleep 2
122	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-result.png
123	```
124
125	4. Stop recording:
126
127	```bash
128	agent-device --session {SESSION} record stop
129	```
130
131	5. Append issue immediately to

People who liked this also viewed

6 picks

Type

dogfood

1.6kstars111forks

View on GitHub

How to install dogfood?

callstackincubator/agent-skills/dogfood

$npx -y skills add callstackincubator/agent-skills --skill dogfood

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Dogfood (agent-device)
2
3	Systematically explore a mobile app, find issues, and produce a report with full reproduction evidence for every finding.
4
5	## Setup
6
7	Only the Target app is required. Everything else has sensible defaults.
8
9	\| Parameter \| Default \| Example override \|
10	\| -------------------- \| ----------------------------------------------------------- \| -------------------------------------------- \|
11	\| Target app \| _(required)_ \| `Settings`, `com.example.app`, deep link URL \|
12	\| Platform \| Infer from user context; otherwise ask (`ios` or `android`) \| `--platform ios` \|
13	\| Session name \| Slugified app/platform (for example `settings-ios`) \| `--session my-session` \|
14	\| Output directory \| `./dogfood-output/` \| `Output directory: /tmp/mobile-qa` \|
15	\| Scope \| Full app \| `Focus on onboarding and profile` \|
16	\| Authentication \| None \| `Sign in to user@example.com` \|
17
18	If the user gives enough context to start, begin immediately with defaults. Ask follow-up only when a required detail is missing (for example platform or credentials).
19
20	Prefer direct `agent-device` binary when available.
21
22	## Workflow
23
24	```
25	1. Initialize Set up session, output dirs, report file
26	2. Launch/Auth Open app and sign in if needed
27	3. Orient Capture initial snapshot and map navigation
28	4. Explore Systematically test flows and states
29	5. Document Record reproducible evidence per issue
30	6. Wrap up Reconcile summary, close session
31	```
32
33	### 1. Initialize
34
35	```bash
36	mkdir -p {OUTPUT_DIR}/screenshots {OUTPUT_DIR}/videos
37	cp {SKILL_DIR}/templates/dogfood-report-template.md {OUTPUT_DIR}/report.md
38	```
39
40	### 2. Launch/Auth
41
42	Start a named session and launch target app:
43
44	```bash
45	agent-device --session {SESSION} open {TARGET_APP} --platform {PLATFORM}
46	agent-device --session {SESSION} snapshot -i
47	```
48
49	If login is required:
50
51	```bash
52	agent-device --session {SESSION} snapshot -i
53	agent-device --session {SESSION} fill @e1 "{EMAIL}"
54	agent-device --session {SESSION} fill @e2 "{PASSWORD}"
55	agent-device --session {SESSION} press @e3
56	agent-device --session {SESSION} wait 1000
57	agent-device --session {SESSION} snapshot -i
58	```
59
60	For OTP/email codes: ask the user, wait for input, then continue.
61
62	### 3. Orient
63
64	Capture initial evidence and navigation anchors:
65
66	```bash
67	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/initial.png
68	agent-device --session {SESSION} snapshot -i
69	```
70
71	Map top-level navigation, tabs, and key workflows before deep testing.
72
73	### 4. Explore
74
75	Read [references/issue-taxonomy.md](references/issue-taxonomy.md) for severity/category calibration.
76
77	Strategy:
78
79	- Move through each major app area (tabs, drawers, settings pages).
80	- Test core journeys end-to-end (create, edit, delete, submit, recover).
81	- Validate edge states (empty/error/loading/offline/permissions denied).
82	- Use `diff snapshot -i` after UI transitions to avoid stale refs.
83	- Periodically capture `logs path` and inspect the app log when behavior looks suspicious.
84
85	Useful commands per screen:
86
87	```bash
88	agent-device --session {SESSION} snapshot -i
89	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/{screen-name}.png
90	agent-device --session {SESSION} appstate
91	agent-device --session {SESSION} logs path
92	```
93
94	### 5. Document Issues (Repro-First)
95
96	Explore and document in one pass. When you find an issue, stop and fully capture evidence before continuing.
97
98	#### Interactive/behavioral issues
99
100	Use video + step screenshots:
101
102	1. Start recording:
103
104	```bash
105	agent-device --session {SESSION} record start {OUTPUT_DIR}/videos/issue-{NNN}-repro.mp4
106	```
107
108	2. Reproduce with visible pacing. Capture each step:
109
110	```bash
111	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-1.png
112	sleep 1
113	# perform action
114	sleep 1
115	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-2.png
116	```
117
118	3. Capture final broken state:
119
120	```bash
121	sleep 2
122	agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-result.png
123	```
124
125	4. Stop recording:
126
127	```bash
128	agent-device --session {SESSION} record stop
129	```
130
131	5. Append issue immediately to

People who liked this also viewed

6 picks

Type