genimg-gemini-web

Name: genimg-gemini-web
Author: proyecto26

30installs2forks

How to install genimg-gemini-web?

proyecto26/sherlock-ai-plugin/genimg-gemini-web

$npx -y skills add proyecto26/sherlock-ai-plugin --skill genimg-gemini-web

Installs into the current project.

Summary

Image generation skill using Gemini Web. Generates images from text prompts via Google Gemini. Also supports text generation. Use as the image generation backend for other skills like cover-image, xhs-images, article-illustrator.

Files · 1

View on GitHub

SKILL.md

1	# Gemini Web Client
2
3	Supports:
4	- Text generation
5	- Image generation (download + save)
6	- Reference image upload (attach images for vision tasks)
7	- Multi-turn conversations within the same executor instance (`keepSession`)
8	- Experimental video generation (`generateVideo`) — Gemini may return an async placeholder; download might require Gemini web UI
9
10	## Quick start
11
12	```bash
13	npx -y bun scripts/main.ts "Hello, Gemini"
14	npx -y bun scripts/main.ts --prompt "Explain quantum computing"
15	npx -y bun scripts/main.ts --prompt "A cute cat" --image cat.png
16	npx -y bun scripts/main.ts --promptfiles system.md content.md --image out.png
17
18	# Multi-turn conversation (agent generates unique sessionId)
19	npx -y bun scripts/main.ts "Remember this: 42" --sessionId my-unique-id-123
20	npx -y bun scripts/main.ts "What number?" --sessionId my-unique-id-123
21	```
22
23	## Executor options (programmatic)
24
25	This skill is typically consumed via `createGeminiWebExecutor(geminiOptions)` (see `scripts/executor.ts`).
26
27	Key options in `GeminiWebOptions`:
28	- `referenceImages?: string \| string[]` Upload local images as references (vision input).
29	- `keepSession?: boolean` Reuse Gemini `chatMetadata` to continue the same conversation across calls (required if you want reference images to persist across multiple messages).
30	- `generateVideo?: string` Generate a video and (best-effort) download to the given path. Gemini may return `video_gen_chip` (async); in that case you must open Gemini web UI to download the result.
31
32	Notes:
33	- `generateVideo` cannot be combined with `generateImage` / `editImage`.
34	- When `keepSession=true` and `referenceImages` is set, reference images are uploaded once per executor instance.
35
36	## Commands
37
38	### Text generation
39
40	```bash
41	# Simple prompt (positional)
42	npx -y bun scripts/main.ts "Your prompt here"
43
44	# Explicit prompt flag
45	npx -y bun scripts/main.ts --prompt "Your prompt here"
46	npx -y bun scripts/main.ts -p "Your prompt here"
47
48	# With model selection
49	npx -y bun scripts/main.ts -p "Hello" -m gemini-2.5-pro
50
51	# Pipe from stdin
52	echo "Summarize this" \| npx -y bun scripts/main.ts
53	```
54
55	### Image generation
56
57	```bash
58	# Generate image with default path (./generated.png)
59	npx -y bun scripts/main.ts --prompt "A sunset over mountains" --image
60
61	# Generate image with custom path
62	npx -y bun scripts/main.ts --prompt "A cute robot" --image robot.png
63
64	# Shorthand
65	npx -y bun scripts/main.ts "A dragon" --image=dragon.png
66	```
67
68	### Output formats
69
70	```bash
71	# Plain text (default)
72	npx -y bun scripts/main.ts "Hello"
73
74	# JSON output
75	npx -y bun scripts/main.ts "Hello" --json
76	```
77
78	## Options
79
80	\| Option \| Description \|
81	\|--------\|-------------\|
82	\| `--prompt <text>`, `-p` \| Prompt text \|
83	\| `--promptfiles <files...>` \| Read prompt from files (concatenated in order) \|
84	\| `--model <id>`, `-m` \| Model: gemini-3-pro (default), gemini-2.5-pro, gemini-2.5-flash \|
85	\| `--image [path]` \| Generate image, save to path (default: generated.png) \|
86	\| `--sessionId <id>` \| Session ID for multi-turn conversation (agent generates unique ID) \|
87	\| `--list-sessions` \| List saved sessions (max 100, sorted by update time) \|
88	\| `--json` \| Output as JSON \|
89	\| `--login` \| Refresh cookies only, then exit \|
90	\| `--cookie-path <path>` \| Custom cookie file path \|
91	\| `--profile-dir <path>` \| Chrome profile directory \|
92	\| `--help`, `-h` \| Show help \|
93
94	CLI note: `scripts/main.ts` supports text generation, image generation, and multi-turn conversations via `--sessionId`. Reference images and video generation are exposed via the executor API.
95
96	## Models
97
98	- `gemini-3-pro` - Default, latest model
99	- `gemini-2.5-pro` - Previous generation pro
100	- `gemini-2.5-flash` - Fast, lightweight
101
102	## Authentication
103
104	First run opens Chrome to authenticate with Google. Cookies are cached for subsequent runs.
105
106	```bash
107	# Force cookie refresh
108	npx -y bun scripts/main.ts --login
109	```
110
111	## Environment variables
112
113	\| Variable \| Description \|
114	\|----------\|-------------\|
115	\| `GEMINI_WEB_DATA_DIR` \| Data directory \|
116	\| `GEMINI_WEB_COOKIE_PATH` \| Cookie file path \|
117	\| `GEMINI_WEB_CHROME_PROFILE_DIR` \| Chrome profile directory \|
118	\| `GEMINI_WEB_CHROME_PATH` \| Chrome executable path \|
119
120	## Examples
121
122	### Generate text response
123	```bash
124	npx -y bun scripts/main.ts "What is the capital of France?"
125	```
126
127	### Generate image
128	```bash
129	npx -y bun scripts/main.ts "A photorealistic image of a golden retriever puppy" --image puppy.png
130	```
131
132	### Get JSON output for parsing
133	```bash
134	npx -y bun scripts/main.ts "Hello" --json \| jq '.text'
135	```
136
137	### Generate image from prompt files
138	```bash
139	# Concatenate system.md + content.md as prompt
140	npx -y bun scripts/main.ts --promptfiles system.md content.md --image output.png
141	```
142
143	### Multi-turn conversation
144	```bash
145	# Start a session with unique ID (agent generates this)
146	npx -y b

People who liked this also viewed

6 picks

Type

genimg-gemini-web

30installs2forks

View on GitHub

How to install genimg-gemini-web?

proyecto26/sherlock-ai-plugin/genimg-gemini-web

$npx -y skills add proyecto26/sherlock-ai-plugin --skill genimg-gemini-web

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Gemini Web Client
2
3	Supports:
4	- Text generation
5	- Image generation (download + save)
6	- Reference image upload (attach images for vision tasks)
7	- Multi-turn conversations within the same executor instance (`keepSession`)
8	- Experimental video generation (`generateVideo`) — Gemini may return an async placeholder; download might require Gemini web UI
9
10	## Quick start
11
12	```bash
13	npx -y bun scripts/main.ts "Hello, Gemini"
14	npx -y bun scripts/main.ts --prompt "Explain quantum computing"
15	npx -y bun scripts/main.ts --prompt "A cute cat" --image cat.png
16	npx -y bun scripts/main.ts --promptfiles system.md content.md --image out.png
17
18	# Multi-turn conversation (agent generates unique sessionId)
19	npx -y bun scripts/main.ts "Remember this: 42" --sessionId my-unique-id-123
20	npx -y bun scripts/main.ts "What number?" --sessionId my-unique-id-123
21	```
22
23	## Executor options (programmatic)
24
25	This skill is typically consumed via `createGeminiWebExecutor(geminiOptions)` (see `scripts/executor.ts`).
26
27	Key options in `GeminiWebOptions`:
28	- `referenceImages?: string \| string[]` Upload local images as references (vision input).
29	- `keepSession?: boolean` Reuse Gemini `chatMetadata` to continue the same conversation across calls (required if you want reference images to persist across multiple messages).
30	- `generateVideo?: string` Generate a video and (best-effort) download to the given path. Gemini may return `video_gen_chip` (async); in that case you must open Gemini web UI to download the result.
31
32	Notes:
33	- `generateVideo` cannot be combined with `generateImage` / `editImage`.
34	- When `keepSession=true` and `referenceImages` is set, reference images are uploaded once per executor instance.
35
36	## Commands
37
38	### Text generation
39
40	```bash
41	# Simple prompt (positional)
42	npx -y bun scripts/main.ts "Your prompt here"
43
44	# Explicit prompt flag
45	npx -y bun scripts/main.ts --prompt "Your prompt here"
46	npx -y bun scripts/main.ts -p "Your prompt here"
47
48	# With model selection
49	npx -y bun scripts/main.ts -p "Hello" -m gemini-2.5-pro
50
51	# Pipe from stdin
52	echo "Summarize this" \| npx -y bun scripts/main.ts
53	```
54
55	### Image generation
56
57	```bash
58	# Generate image with default path (./generated.png)
59	npx -y bun scripts/main.ts --prompt "A sunset over mountains" --image
60
61	# Generate image with custom path
62	npx -y bun scripts/main.ts --prompt "A cute robot" --image robot.png
63
64	# Shorthand
65	npx -y bun scripts/main.ts "A dragon" --image=dragon.png
66	```
67
68	### Output formats
69
70	```bash
71	# Plain text (default)
72	npx -y bun scripts/main.ts "Hello"
73
74	# JSON output
75	npx -y bun scripts/main.ts "Hello" --json
76	```
77
78	## Options
79
80	\| Option \| Description \|
81	\|--------\|-------------\|
82	\| `--prompt <text>`, `-p` \| Prompt text \|
83	\| `--promptfiles <files...>` \| Read prompt from files (concatenated in order) \|
84	\| `--model <id>`, `-m` \| Model: gemini-3-pro (default), gemini-2.5-pro, gemini-2.5-flash \|
85	\| `--image [path]` \| Generate image, save to path (default: generated.png) \|
86	\| `--sessionId <id>` \| Session ID for multi-turn conversation (agent generates unique ID) \|
87	\| `--list-sessions` \| List saved sessions (max 100, sorted by update time) \|
88	\| `--json` \| Output as JSON \|
89	\| `--login` \| Refresh cookies only, then exit \|
90	\| `--cookie-path <path>` \| Custom cookie file path \|
91	\| `--profile-dir <path>` \| Chrome profile directory \|
92	\| `--help`, `-h` \| Show help \|
93
94	CLI note: `scripts/main.ts` supports text generation, image generation, and multi-turn conversations via `--sessionId`. Reference images and video generation are exposed via the executor API.
95
96	## Models
97
98	- `gemini-3-pro` - Default, latest model
99	- `gemini-2.5-pro` - Previous generation pro
100	- `gemini-2.5-flash` - Fast, lightweight
101
102	## Authentication
103
104	First run opens Chrome to authenticate with Google. Cookies are cached for subsequent runs.
105
106	```bash
107	# Force cookie refresh
108	npx -y bun scripts/main.ts --login
109	```
110
111	## Environment variables
112
113	\| Variable \| Description \|
114	\|----------\|-------------\|
115	\| `GEMINI_WEB_DATA_DIR` \| Data directory \|
116	\| `GEMINI_WEB_COOKIE_PATH` \| Cookie file path \|
117	\| `GEMINI_WEB_CHROME_PROFILE_DIR` \| Chrome profile directory \|
118	\| `GEMINI_WEB_CHROME_PATH` \| Chrome executable path \|
119
120	## Examples
121
122	### Generate text response
123	```bash
124	npx -y bun scripts/main.ts "What is the capital of France?"
125	```
126
127	### Generate image
128	```bash
129	npx -y bun scripts/main.ts "A photorealistic image of a golden retriever puppy" --image puppy.png
130	```
131
132	### Get JSON output for parsing
133	```bash
134	npx -y bun scripts/main.ts "Hello" --json \| jq '.text'
135	```
136
137	### Generate image from prompt files
138	```bash
139	# Concatenate system.md + content.md as prompt
140	npx -y bun scripts/main.ts --promptfiles system.md content.md --image output.png
141	```
142
143	### Multi-turn conversation
144	```bash
145	# Start a session with unique ID (agent generates this)
146	npx -y b

People who liked this also viewed

6 picks

Type