@monoes/monomindcli 1.10.31 → 1.10.33

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.
@@ -13,21 +13,20 @@ triggers:
13
13
  tools:
14
14
  - Bash
15
15
  requires:
16
- - agent-browser >= 0.25.4
16
+ - monomind >= 1.0.0
17
17
  ---
18
18
 
19
- <!-- Pattern adapted from agent-browser — rebranded for monomind -->
20
19
 
21
20
  # Electron App Automation (monomind:browse-electron)
22
21
 
23
- Automate any Electron desktop app using agent-browser via Chrome DevTools Protocol (CDP). Electron apps are built on Chromium and expose a CDP port, enabling the same snapshot-interact workflow used for web pages.
22
+ Automate any Electron desktop app using monomind browse via Chrome DevTools Protocol (CDP). Electron apps are built on Chromium and expose a CDP port, enabling the same snapshot-interact workflow used for web pages.
24
23
 
25
24
  See `monomind:browse` for the full browser automation reference.
26
25
 
27
26
  ## Core Workflow
28
27
 
29
28
  1. **Launch** the Electron app with remote debugging enabled
30
- 2. **Connect** agent-browser to the CDP port
29
+ 2. **Connect** monomind browse to the CDP port
31
30
  3. **Snapshot** to discover interactive elements
32
31
  4. **Interact** using element refs
33
32
  5. **Re-snapshot** after navigation or state changes
@@ -36,13 +35,13 @@ See `monomind:browse` for the full browser automation reference.
36
35
  # Launch an Electron app with remote debugging
37
36
  open -a "Slack" --args --remote-debugging-port=9222
38
37
 
39
- # Connect agent-browser to the app
40
- agent-browser connect 9222
38
+ # Connect monomind browse to the app
39
+ npx monomind browse open --port 0
41
40
 
42
41
  # Standard workflow from here
43
- agent-browser snapshot -i
44
- agent-browser click @e5
45
- agent-browser screenshot slack-desktop.png
42
+ npx monomind browse snapshot -i
43
+ npx monomind browse click @e5
44
+ npx monomind browse screenshot slack-desktop.png
46
45
  ```
47
46
 
48
47
  ## Launching with CDP
@@ -81,13 +80,13 @@ discord --remote-debugging-port=9224
81
80
 
82
81
  ```bash
83
82
  # Connect to a specific port (persists for session)
84
- agent-browser connect 9222
83
+ npx monomind browse open --port 0
85
84
 
86
85
  # Or pass --cdp on each command
87
- agent-browser --cdp 9222 snapshot -i
86
+ npx monomind browse --cdp 9222 snapshot -i
88
87
 
89
88
  # Auto-discover a running Chromium-based app
90
- agent-browser --auto-connect snapshot -i
89
+ npx monomind browse --auto-connect snapshot -i
91
90
  ```
92
91
 
93
92
  ## Tab Management
@@ -95,9 +94,9 @@ agent-browser --auto-connect snapshot -i
95
94
  Electron apps often have multiple windows or webviews:
96
95
 
97
96
  ```bash
98
- agent-browser tab # list all targets (windows, webviews)
99
- agent-browser tab t2 # switch by stable id
100
- agent-browser tab --url "*settings*" # switch by URL pattern
97
+ npx monomind browse tab # list all targets (windows, webviews)
98
+ npx monomind browse tab t2 # switch by stable id
99
+ npx monomind browse tab --url "*settings*" # switch by URL pattern
101
100
  ```
102
101
 
103
102
  ## Webview Support
@@ -105,14 +104,14 @@ agent-browser tab --url "*settings*" # switch by URL pattern
105
104
  Electron `<webview>` elements appear as separate targets in the tab list:
106
105
 
107
106
  ```bash
108
- agent-browser connect 9222
109
- agent-browser tab
107
+ npx monomind browse open --port 0
108
+ npx monomind browse tab
110
109
  # 0: [page] Slack - Main Window https://app.slack.com/
111
110
  # 1: [webview] Embedded Content https://example.com/widget
112
111
 
113
- agent-browser tab t2 # switch to webview
114
- agent-browser snapshot -i
115
- agent-browser click @e3
112
+ npx monomind browse tab t2 # switch to webview
113
+ npx monomind browse snapshot -i
114
+ npx monomind browse click @e3
116
115
  ```
117
116
 
118
117
  ## Common Patterns
@@ -122,57 +121,57 @@ agent-browser click @e3
122
121
  ```bash
123
122
  open -a "Slack" --args --remote-debugging-port=9222
124
123
  sleep 3
125
- agent-browser connect 9222
126
- agent-browser snapshot -i
127
- agent-browser click @e10
128
- agent-browser snapshot -i
124
+ npx monomind browse open --port 0
125
+ npx monomind browse snapshot -i
126
+ npx monomind browse click @e10
127
+ npx monomind browse snapshot -i
129
128
  ```
130
129
 
131
130
  ### Screenshots
132
131
 
133
132
  ```bash
134
- agent-browser connect 9222
135
- agent-browser screenshot app-state.png
136
- agent-browser screenshot --annotate annotated-app.png
133
+ npx monomind browse open --port 0
134
+ npx monomind browse screenshot app-state.png
135
+ npx monomind browse screenshot --annotate annotated-app.png
137
136
  ```
138
137
 
139
138
  ### Extract Data
140
139
 
141
140
  ```bash
142
- agent-browser connect 9222
143
- agent-browser snapshot -i
144
- agent-browser get text @e5
145
- agent-browser snapshot --json > app-state.json
141
+ npx monomind browse open --port 0
142
+ npx monomind browse snapshot -i
143
+ npx monomind browse get text @e5
144
+ npx monomind browse snapshot --json > app-state.json
146
145
  ```
147
146
 
148
147
  ### Fill Forms
149
148
 
150
149
  ```bash
151
- agent-browser connect 9222
152
- agent-browser snapshot -i
153
- agent-browser fill @e3 "search query"
154
- agent-browser press Enter
155
- agent-browser wait 1000
156
- agent-browser snapshot -i
150
+ npx monomind browse open --port 0
151
+ npx monomind browse snapshot -i
152
+ npx monomind browse fill @e3 "search query"
153
+ npx monomind browse press Enter
154
+ npx monomind browse wait 1000
155
+ npx monomind browse snapshot -i
157
156
  ```
158
157
 
159
158
  ### Multiple Apps Simultaneously
160
159
 
161
160
  ```bash
162
- agent-browser --session slack connect 9222
163
- agent-browser --session vscode connect 9223
161
+ npx monomind browse --session slack connect 9222
162
+ npx monomind browse --session vscode connect 9223
164
163
 
165
- agent-browser --session slack snapshot -i
166
- agent-browser --session vscode snapshot -i
164
+ npx monomind browse --session slack snapshot -i
165
+ npx monomind browse --session vscode snapshot -i
167
166
  ```
168
167
 
169
168
  ## Color Scheme
170
169
 
171
170
  ```bash
172
- agent-browser connect 9222
173
- agent-browser --color-scheme dark snapshot -i
171
+ npx monomind browse open --port 0
172
+ npx monomind browse --color-scheme dark snapshot -i
174
173
  # or
175
- AGENT_BROWSER_COLOR_SCHEME=dark agent-browser connect 9222
174
+ AGENT_BROWSER_COLOR_SCHEME=dark npx monomind browse open --port 0
176
175
  ```
177
176
 
178
177
  ## Troubleshooting
@@ -181,8 +180,8 @@ AGENT_BROWSER_COLOR_SCHEME=dark agent-browser connect 9222
181
180
  |---------|-----|
182
181
  | "Connection refused" | App not launched with `--remote-debugging-port`, or port mismatch |
183
182
  | Connect fails after launch | Wait a few seconds — `sleep 3` before connect |
184
- | Elements not in snapshot | Use `agent-browser tab` to switch to the right window/webview |
185
- | Can't type in inputs | Try `agent-browser keyboard type "text"` or `keyboard inserttext "text"` |
183
+ | Elements not in snapshot | Use `monomind browse tab` to switch to the right window/webview |
184
+ | Can't type in inputs | Try `monomind browse keyboard type "text"` or `keyboard inserttext "text"` |
186
185
 
187
186
  ## Supported Apps
188
187
 
@@ -16,10 +16,9 @@ triggers:
16
16
  tools:
17
17
  - Bash
18
18
  requires:
19
- - agent-browser >= 0.25.4
19
+ - monomind >= 1.0.0
20
20
  ---
21
21
 
22
- <!-- Pattern adapted from agent-browser — rebranded for monomind -->
23
22
 
24
23
  # Systematic QA / Dogfood Testing (monomind:browse-qa)
25
24
 
@@ -39,7 +38,7 @@ See `monomind:browse` for the full browser automation reference.
39
38
 
40
39
  Start immediately with defaults. Only ask if authentication is mentioned but credentials are missing.
41
40
 
42
- Always call `agent-browser` directly — never `npx agent-browser`. The direct binary uses the fast Rust client.
41
+ Always call `monomind browse` directly — using `npx monomind browse`. It uses the native TypeScript CDP client.
43
42
 
44
43
  ## Workflow
45
44
 
@@ -57,21 +56,21 @@ Always call `agent-browser` directly — never `npx agent-browser`. The direct b
57
56
  ```bash
58
57
  mkdir -p ./qa-output/screenshots
59
58
  SESSION="$(echo '<TARGET_URL>' | sed 's|https\?://||; s|/.*||; s|\.|-|g')"
60
- agent-browser --session "$SESSION" open <TARGET_URL>
61
- agent-browser --session "$SESSION" wait --load networkidle
59
+ npx monomind browse --session "$SESSION" open <TARGET_URL>
60
+ npx monomind browse --session "$SESSION" wait --load networkidle
62
61
  ```
63
62
 
64
63
  ### 2. Authenticate (if needed)
65
64
 
66
65
  ```bash
67
- agent-browser --session "$SESSION" snapshot -i
68
- agent-browser --session "$SESSION" fill @e1 "<EMAIL>"
69
- agent-browser --session "$SESSION" fill @e2 "<PASSWORD>"
70
- agent-browser --session "$SESSION" click @e3
71
- agent-browser --session "$SESSION" wait --load networkidle
66
+ npx monomind browse --session "$SESSION" snapshot -i
67
+ npx monomind browse --session "$SESSION" fill @e1 "<EMAIL>"
68
+ npx monomind browse --session "$SESSION" fill @e2 "<PASSWORD>"
69
+ npx monomind browse --session "$SESSION" click @e3
70
+ npx monomind browse --session "$SESSION" wait --load networkidle
72
71
 
73
72
  # Save auth state for reuse
74
- agent-browser --session "$SESSION" state save ./qa-output/auth-state.json
73
+ npx monomind browse --session "$SESSION" state save ./qa-output/auth-state.json
75
74
  ```
76
75
 
77
76
  For OTP/email codes: ask the user, wait for input, then enter the code.
@@ -79,10 +78,10 @@ For OTP/email codes: ask the user, wait for input, then enter the code.
79
78
  ### 3. Orient
80
79
 
81
80
  ```bash
82
- agent-browser --session "$SESSION" screenshot --annotate ./qa-output/screenshots/00-initial.png
83
- agent-browser --session "$SESSION" snapshot -i
84
- agent-browser --session "$SESSION" get title
85
- agent-browser --session "$SESSION" get url
81
+ npx monomind browse --session "$SESSION" screenshot --annotate ./qa-output/screenshots/00-initial.png
82
+ npx monomind browse --session "$SESSION" snapshot -i
83
+ npx monomind browse --session "$SESSION" get title
84
+ npx monomind browse --session "$SESSION" get url
86
85
  ```
87
86
 
88
87
  Document the starting state: what is visible, main navigation elements, key actions available.
@@ -92,17 +91,17 @@ Document the starting state: what is visible, main navigation elements, key acti
92
91
  For each area/page:
93
92
 
94
93
  ```bash
95
- agent-browser --session "$SESSION" click @e[nav-item]
96
- agent-browser --session "$SESSION" wait --load networkidle
97
- agent-browser --session "$SESSION" snapshot -i
98
- agent-browser --session "$SESSION" screenshot ./qa-output/screenshots/<page-name>.png
94
+ npx monomind browse --session "$SESSION" click @e[nav-item]
95
+ npx monomind browse --session "$SESSION" wait --load networkidle
96
+ npx monomind browse --session "$SESSION" snapshot -i
97
+ npx monomind browse --session "$SESSION" screenshot ./qa-output/screenshots/<page-name>.png
99
98
  ```
100
99
 
101
100
  **Test systematically:**
102
101
  - Navigation: does every link work? Does back/forward work?
103
102
  - Forms: submit empty, submit invalid, submit valid
104
103
  - Interactive elements: hover states, click states, disabled states
105
- - Responsive: `agent-browser set viewport 375 812` (mobile), `1280 720` (desktop)
104
+ - Responsive: `monomind browse set viewport 375 812` (mobile), `1280 720` (desktop)
106
105
  - Error states: what happens when APIs fail? Use `network route` to simulate failures
107
106
  - Loading states: `wait --load networkidle` then check if spinners resolve
108
107
 
@@ -112,7 +111,7 @@ For every issue found:
112
111
 
113
112
  ```bash
114
113
  # 1. Screenshot the broken state
115
- agent-browser --session "$SESSION" screenshot ./qa-output/screenshots/bug-<N>-<name>.png
114
+ npx monomind browse --session "$SESSION" screenshot ./qa-output/screenshots/bug-<N>-<name>.png
116
115
 
117
116
  # 2. Record reproduction steps in report
118
117
  ```
@@ -134,7 +133,7 @@ Issue format:
134
133
  ### 6. Wrap Up
135
134
 
136
135
  ```bash
137
- agent-browser --session "$SESSION" close
136
+ npx monomind browse --session "$SESSION" close
138
137
  ```
139
138
 
140
139
  Print summary:
@@ -177,12 +176,12 @@ Output: ./qa-output/
177
176
  - [ ] Tooltips appear on hover
178
177
 
179
178
  **Responsive:**
180
- - [ ] Mobile (375×812): `agent-browser set viewport 375 812`
181
- - [ ] Tablet (768×1024): `agent-browser set viewport 768 1024`
182
- - [ ] Desktop (1280×720): `agent-browser set viewport 1280 720`
179
+ - [ ] Mobile (375×812): `monomind browse set viewport 375 812`
180
+ - [ ] Tablet (768×1024): `monomind browse set viewport 768 1024`
181
+ - [ ] Desktop (1280×720): `monomind browse set viewport 1280 720`
183
182
 
184
183
  **Error states:**
185
- - [ ] Network error: `agent-browser network route "https://api.*" --abort`
184
+ - [ ] Network error: `monomind browse network route "https://api.*" --abort`
186
185
  - [ ] Empty state: check when lists/tables have no data
187
186
  - [ ] 404 page: navigate to `/nonexistent-path`
188
187
 
@@ -195,11 +194,11 @@ Output: ./qa-output/
195
194
 
196
195
  ```bash
197
196
  # Block all API calls to test error handling
198
- agent-browser batch \
197
+ npx monomind browse batch \
199
198
  '["open"]' \
200
199
  '["network", "route", "https://api.example.com/*", "--abort"]' \
201
200
  '["navigate", "https://app.example.com/dashboard"]'
202
- agent-browser snapshot -i
201
+ npx monomind browse snapshot -i
203
202
  # → Verify error states render correctly
204
203
  ```
205
204
 
@@ -207,10 +206,10 @@ agent-browser snapshot -i
207
206
 
208
207
  ```bash
209
208
  # Baseline
210
- agent-browser open https://app.example.com && agent-browser snapshot -i > ./qa-output/baseline.txt
209
+ npx monomind browse open https://app.example.com && monomind browse snapshot -i > ./qa-output/baseline.txt
211
210
 
212
211
  # After change
213
- agent-browser open https://app.example.com && agent-browser diff snapshot --baseline ./qa-output/baseline.txt
212
+ npx monomind browse open https://app.example.com && monomind browse diff snapshot --baseline ./qa-output/baseline.txt
214
213
  ```
215
214
 
216
215
  ## Monomind Integration
@@ -1,6 +1,5 @@
1
1
  # Authentication Patterns
2
2
 
3
- <!-- Pattern adapted from agent-browser — rebranded for monomind -->
4
3
 
5
4
  Login flows, session persistence, OAuth, 2FA, and authenticated browsing.
6
5
 
@@ -27,13 +26,13 @@ Log in to your target site normally in this Chrome window.
27
26
  **Step 2: Save auth state**
28
27
 
29
28
  ```bash
30
- agent-browser --auto-connect state save ./my-auth.json
29
+ npx monomind browse --auto-connect state save ./my-auth.json
31
30
  ```
32
31
 
33
32
  **Step 3: Reuse in automation**
34
33
 
35
34
  ```bash
36
- agent-browser --state ./my-auth.json open https://app.example.com/dashboard
35
+ npx monomind browse --state ./my-auth.json open https://app.example.com/dashboard
37
36
  ```
38
37
 
39
38
  ## Auth Methods Quick Reference
@@ -51,24 +50,24 @@ agent-browser --state ./my-auth.json open https://app.example.com/dashboard
51
50
  ## Basic Login Flow
52
51
 
53
52
  ```bash
54
- agent-browser open https://app.example.com/login
55
- agent-browser snapshot -i
53
+ npx monomind browse open https://app.example.com/login
54
+ npx monomind browse snapshot -i
56
55
  # identify: @e[email], @e[password], @e[submit]
57
- agent-browser fill @e1 "user@test.com"
58
- agent-browser fill @e2 "SecurePass123!"
59
- agent-browser click @e3
60
- agent-browser wait --url "**/dashboard" --timeout 10000
56
+ npx monomind browse fill @e1 "user@test.com"
57
+ npx monomind browse fill @e2 "SecurePass123!"
58
+ npx monomind browse click @e3
59
+ npx monomind browse wait --url "**/dashboard" --timeout 10000
61
60
 
62
61
  # Save for reuse
63
- agent-browser state save ./auth.json
62
+ npx monomind browse state save ./auth.json
64
63
  ```
65
64
 
66
65
  ## Session Persistence
67
66
 
68
67
  ```bash
69
68
  # Auto-save/restore state by name
70
- agent-browser --session-name myapp open https://app.example.com
71
- # → State saved to ~/.agent-browser/sessions/myapp automatically
69
+ npx monomind browse --session-name myapp open https://app.example.com
70
+ # → State saved to ~/.monomind/browser-sessions/myapp automatically
72
71
  # → Next time: just --session-name myapp and you're logged in
73
72
  ```
74
73
 
@@ -77,7 +76,7 @@ agent-browser --session-name myapp open https://app.example.com
77
76
  ```bash
78
77
  # Generate a key: openssl rand -hex 32
79
78
  export AGENT_BROWSER_ENCRYPTION_KEY=<64-char-hex>
80
- agent-browser --session-name secure open https://app.example.com
79
+ npx monomind browse --session-name secure open https://app.example.com
81
80
  # → State file is AES-256-GCM encrypted at rest
82
81
  ```
83
82
 
@@ -85,28 +84,28 @@ agent-browser --session-name secure open https://app.example.com
85
84
 
86
85
  ```bash
87
86
  # Store credentials once
88
- echo "mypassword" | agent-browser auth save github \
87
+ echo "mypassword" | monomind browse auth save github \
89
88
  --url https://github.com/login \
90
89
  --username me \
91
90
  --password-stdin
92
91
 
93
92
  # Login using stored credentials
94
- agent-browser auth login github
93
+ npx monomind browse auth login github
95
94
  ```
96
95
 
97
96
  ## Token-Based Auth (Skip Login Form)
98
97
 
99
98
  ```bash
100
99
  # Headers scoped to origin only — never leaked to other domains
101
- agent-browser open https://api.example.com --headers '{"Authorization": "Bearer <token>"}'
102
- agent-browser snapshot -i
100
+ npx monomind browse open https://api.example.com --headers '{"Authorization": "Bearer <token>"}'
101
+ npx monomind browse snapshot -i
103
102
  ```
104
103
 
105
104
  ## Cookie Import (from cURL)
106
105
 
107
106
  ```bash
108
107
  # Export from browser DevTools → Network → Copy as cURL → save to file
109
- agent-browser cookies set --curl ./cookies.curl # auto-detects format
108
+ npx monomind browse cookies set --curl ./cookies.curl # auto-detects format
110
109
  ```
111
110
 
112
111
  ## OAuth / SSO
@@ -114,43 +113,43 @@ agent-browser cookies set --curl ./cookies.curl # auto-detects format
114
113
  OAuth flows redirect through external providers — just follow the snapshots:
115
114
 
116
115
  ```bash
117
- agent-browser open https://app.example.com/login
118
- agent-browser snapshot -i
119
- agent-browser click @e[sign-in-with-google]
120
- agent-browser wait --url "**/accounts.google.com/**"
121
- agent-browser snapshot -i
116
+ npx monomind browse open https://app.example.com/login
117
+ npx monomind browse snapshot -i
118
+ npx monomind browse click @e[sign-in-with-google]
119
+ npx monomind browse wait --url "**/accounts.google.com/**"
120
+ npx monomind browse snapshot -i
122
121
  # fill Google credentials...
123
- agent-browser wait --url "**/app.example.com/**"
122
+ npx monomind browse wait --url "**/app.example.com/**"
124
123
  # → OAuth complete, save state
125
- agent-browser state save ./oauth-auth.json
124
+ npx monomind browse state save ./oauth-auth.json
126
125
  ```
127
126
 
128
127
  ## Two-Factor Authentication
129
128
 
130
129
  ```bash
131
- agent-browser open https://app.example.com/login
132
- agent-browser fill @e1 "user@test.com"
133
- agent-browser fill @e2 "password"
134
- agent-browser click @e3
135
- agent-browser wait --text "Enter your code"
130
+ npx monomind browse open https://app.example.com/login
131
+ npx monomind browse fill @e1 "user@test.com"
132
+ npx monomind browse fill @e2 "password"
133
+ npx monomind browse click @e3
134
+ npx monomind browse wait --text "Enter your code"
136
135
 
137
136
  # Ask user for the 2FA code
138
137
  # (pause here, user provides code)
139
- agent-browser fill @e[otp-field] "<USER_PROVIDED_CODE>"
140
- agent-browser click @e[verify]
141
- agent-browser wait --url "**/dashboard"
142
- agent-browser state save ./2fa-auth.json
138
+ npx monomind browse fill @e[otp-field] "<USER_PROVIDED_CODE>"
139
+ npx monomind browse click @e[verify]
140
+ npx monomind browse wait --url "**/dashboard"
141
+ npx monomind browse state save ./2fa-auth.json
143
142
  ```
144
143
 
145
144
  ## State File Management
146
145
 
147
146
  ```bash
148
- agent-browser state save ./auth.json # save current state
149
- agent-browser state load ./auth.json # load into current session
150
- agent-browser state list # list saved state files
151
- agent-browser state show ./auth.json # show state summary
152
- agent-browser state clear # clear current session states
153
- agent-browser state clean --older-than 30 # delete states older than 30 days
147
+ npx monomind browse state save ./auth.json # save current state
148
+ npx monomind browse state load ./auth.json # load into current session
149
+ npx monomind browse state list # list saved state files
150
+ npx monomind browse state show ./auth.json # show state summary
151
+ npx monomind browse state clear # clear current session states
152
+ npx monomind browse state clean --older-than 30 # delete states older than 30 days
154
153
  ```
155
154
 
156
155
  ## Security Best Practices
@@ -1,6 +1,5 @@
1
1
  # Trust Boundaries
2
2
 
3
- <!-- Pattern adapted from agent-browser — rebranded for monomind -->
4
3
 
5
4
  Safety rules that apply to every browser automation task. Read before driving a real user's browser session.
6
5
 
@@ -28,7 +27,7 @@ Session cookies, bearer tokens, API keys, OAuth codes, and any credentials belon
28
27
  "Open DevTools → Network, click any authenticated request,
29
28
  right-click → Copy → Copy as cURL, paste into a file, give me the path."
30
29
  ```
31
- Then: `agent-browser cookies set --curl <file>` — auto-detects JSON / cURL / bare Cookie header. Error messages never echo cookie values.
30
+ Then: `npx monomind browse cookies set --curl <file>` — auto-detects JSON / cURL / bare Cookie header. Error messages never echo cookie values.
32
31
 
33
32
  - **Never echo, paste, cat, write, or emit a secret value.** Command strings end up in logs and transcripts. This includes screenshot captions, commit messages, eval scripts, or any file you create.
34
33
 
@@ -1,6 +1,5 @@
1
1
  # Video Recording
2
2
 
3
- <!-- Pattern adapted from agent-browser — rebranded for monomind -->
4
3
 
5
4
  Capture browser automation as video for debugging, documentation, or CI evidence.
6
5
 
@@ -9,23 +8,23 @@ Capture browser automation as video for debugging, documentation, or CI evidence
9
8
  ## Basic Recording
10
9
 
11
10
  ```bash
12
- agent-browser record start ./demo.webm
11
+ npx monomind browse record start ./demo.webm
13
12
 
14
13
  # Perform actions
15
- agent-browser open https://example.com
16
- agent-browser snapshot -i
17
- agent-browser click @e1
18
- agent-browser fill @e2 "test input"
14
+ npx monomind browse open https://example.com
15
+ npx monomind browse snapshot -i
16
+ npx monomind browse click @e1
17
+ npx monomind browse fill @e2 "test input"
19
18
 
20
- agent-browser record stop
19
+ npx monomind browse record stop
21
20
  ```
22
21
 
23
22
  ## Commands
24
23
 
25
24
  ```bash
26
- agent-browser record start ./output.webm # start recording to file
27
- agent-browser record stop # stop current recording
28
- agent-browser record restart ./take2.webm # stop current + start new
25
+ npx monomind browse record start ./output.webm # start recording to file
26
+ npx monomind browse record stop # stop current recording
27
+ npx monomind browse record restart ./take2.webm # stop current + start new
29
28
  ```
30
29
 
31
30
  ## Patterns
@@ -33,38 +32,38 @@ agent-browser record restart ./take2.webm # stop current + start new
33
32
  ### Debug a failed automation
34
33
 
35
34
  ```bash
36
- agent-browser record start ./debug-$(date +%Y%m%d-%H%M%S).webm
35
+ npx monomind browse record start ./debug-$(date +%Y%m%d-%H%M%S).webm
37
36
 
38
- agent-browser open https://app.example.com
39
- agent-browser snapshot -i
40
- agent-browser click @e1 || { echo "Click failed"; agent-browser record stop; exit 1; }
37
+ npx monomind browse open https://app.example.com
38
+ npx monomind browse snapshot -i
39
+ npx monomind browse click @e1 || { echo "Click failed"; monomind browse record stop; exit 1; }
41
40
 
42
- agent-browser record stop
41
+ npx monomind browse record stop
43
42
  ```
44
43
 
45
44
  ### CI/CD test evidence
46
45
 
47
46
  ```bash
48
47
  mkdir -p ./test-recordings
49
- agent-browser record start ./test-recordings/e2e-$(date +%s).webm
48
+ npx monomind browse record start ./test-recordings/e2e-$(date +%s).webm
50
49
 
51
50
  # run tests...
52
51
 
53
- agent-browser record stop
52
+ npx monomind browse record stop
54
53
  ```
55
54
 
56
55
  ### Combine with screenshots
57
56
 
58
57
  ```bash
59
- agent-browser record start ./flow.webm
58
+ npx monomind browse record start ./flow.webm
60
59
 
61
- agent-browser open https://example.com
62
- agent-browser screenshot ./screenshots/step1.png
60
+ npx monomind browse open https://example.com
61
+ npx monomind browse screenshot ./screenshots/step1.png
63
62
 
64
- agent-browser click @e1
65
- agent-browser screenshot ./screenshots/step2.png
63
+ npx monomind browse click @e1
64
+ npx monomind browse screenshot ./screenshots/step2.png
66
65
 
67
- agent-browser record stop
66
+ npx monomind browse record stop
68
67
  ```
69
68
 
70
69
  ## Best Practices
@@ -73,7 +72,7 @@ agent-browser record stop
73
72
  2. **Descriptive filenames** — include date/context: `login-flow-2026-05-17.webm`
74
73
  3. **Trap cleanup** — always stop recording on exit:
75
74
  ```bash
76
- cleanup() { agent-browser record stop 2>/dev/null || true; }
75
+ cleanup() { monomind browse record stop 2>/dev/null || true; }
77
76
  trap cleanup EXIT
78
77
  ```
79
78