promptcmd

Name: promptcmd
Author: tgalal

48stars2forks

How to install promptcmd?

tgalal/promptcmd

$git clone https://github.com/tgalal/promptcmd

Installs into the current project.

Summary

Quick Start | Documentation | Examples

Files · 1

View on GitHub

README.md

1	# promptcmd
2
3	[Quick Start](#quick-start) \| [Documentation](https://docs.promptcmd.sh) \| [Examples](#examples)
4
5	promptcmd is a manager and executor for programmable prompts. Define a
6	prompt template once, then execute it like any other terminal command, complete
7	with argument parsing, `--help` text, and stdin/stdout integration.
8
9	![promptcmd demo](./docs/img/create.gif)
10
11	Unlike tools that require you to manually manage prompt files or rely on
12	implicit tool-calling, promptcmd gives you explicit control over what data your
13	models access. Build compositional workflows by nesting prompts, executing
14	shell commands within templates, and piping data through multi-step AI
15	pipelines.
16
17	## Key Features
18
19	### Prompts as CLI Commands
20
21	Create a `.prompt` file, enable it with `promptctl`, and execute it like any
22	native tool.
23
24	```bash
25	$ promptctl create bashme_a_script_that
26	$ bashme_a_script_that renames all files in current directly to ".backup"
27	```
28
29	[More on Execution](https://docs.promptcmd.sh/usage/exec).
30
31	### Execute in Remote Shells
32
33	Prepend SSH commands with `promptctl`, your prompts magically appear in your
34	remote shell sessions.
35
36	```bash
37	$ promptctl ssh user@server
38	server $ bashme_a_script_that renames all files in current directly to ".backup"
39	```
40
41	[More on SSH Integration](https://docs.promptcmd.sh/integrations/ssh).
42
43	### Local and Remote Provider Support
44
45	Use your Ollama endpoint or configure an API key for OpenAI, OpenRouter,
46	Anthropic, Google, or MiniMax. Swap between them with ease.
47
48	```bash
49	$ promptctl create render-md
50	$ cat README.md \| render-md -m openai
51	$ cat README.md \| render-md -m ollama/gpt-oss:20b
52	```
53
54	[More on Providers](https://docs.promptcmd.sh/configuration/providers).
55
56	### Group and Load Balancing
57
58	Distribute requests across several providers with equal or weighted distribution
59	for cost optimization.
60
61	```toml
62	# config.toml
63	[groups.balanced]
64	providers = ["openai", "google"]
65	```
66	```bash
67	$ cat README.md \| render-md -m balanced
68	```
69
70	[More on Groups](https://docs.promptcmd.sh/configuration/groups).
71
72	### Caching
73
74	Cache responses for a configured amount of time for adding determinism in
75	pipelines and more efficient token consumption.
76
77	```toml
78	# config.toml
79	[providers.openai]
80	cache_ttl = 60 # number of seconds
81	```
82
83	Set/Override during execution:
84
85	```bash
86	$ cat README.md \| render-md --config-cache-ttl 120
87	```
88
89	[More on Caching](https://docs.promptcmd.sh/configuration/caching).
90
91	### Custom Models with Character
92
93	Use Variants to define custom models with own personality or specialization in
94	tasks:
95
96	```config.toml
97	[providers.anthropic]
98	api_key = "sk-xxxxx"
99	model = "claude-sonnet-4-5"
100
101	[providers.anthropic.glados]
102	system = "Use sarcasm and offending jokes like the GlaDoS character from Portal."
103
104	[providers.anthropic.wheatley]
105	system = "Reply as if you are Wheatley from Portal."
106	```
107
108	```bash
109	$ tipoftheday -m glados
110	$ tipoftheday -m wheatley
111	```
112
113	[More on Variants](https://docs.promptcmd.sh/configuration/variants).
114
115	## Quick Start
116
117	### Install
118
119	Linux/macOS:
120	```bash
121	curl -LsSf https://installer.promptcmd.sh \| sh
122	```
123
124	macOS (Homebrew):
125	```bash
126	brew install tgalal/tap/promptcmd
127	```
128
129	Windows (PowerShell):
130	```powershell
131	powershell -ExecutionPolicy Bypass -c "irm https://installer-ps.promptcmd.sh \| iex"
132	```
133
134	### Configure API Keys
135
136	Configure your API keys by editing `config.toml`:
137
138	```bash
139	promptctl config edit
140	```
141	Find your provider's name, e.g., for anthropic:
142
143	```toml
144	[providers.anthropic]
145	api_key = "sk-ant-api03-..."
146	```
147
148	Alternatively, you can set the keys via Environment Variables:
149
150	```
151	PROMPTCMD_ANTHROPIC_API_KEY="your_api_key"
152	PROMPTCMD_OPENAI_API_KEY="your_api_key"
153	PROMPTCMD_ANTHROPIC_API_KEY="your_api_key"
154	PROMPTCMD_OPENROUTER_API_KEY="your_api_key"
155	PROMPTCMD_MINIMAX_API_KEY="your_api_key"
156	```
157
158	### Create Your First Prompt
159
160	Create a `summarize.prompt` file:
161
162	```bash
163	promptctl create summarize
164	```
165
166	Insert the following:
167
168	```yaml
169	---
170	model: anthropic/claude-sonnet-4-5
171	input:
172	schema:
173	words?: integer, Summary

People who liked this also viewed

6 picks

Type

bun-promptxTerminal prompt library for the Bun runtimePromptsSep 20232.6k146
lintlanglintlang is a static linter for AI agent configs, tool descriptions, and system prompts that runs zero-LLM quality gating in CI. Catches language-level failures (vague tool descriptions, missing stop conditions, schema gaps) before they reach runtime, with deterministic regex + structural detectors and no model calls.PromptsJul 202681152
gpt-cost-estimatorA cost estimator for OpenAI API calls in tqdm loops.PromptsNov 202415120
shadowquillappShadowQuill is a privacy‑first, offline and free prompt engineering/generation app powered by self-hosted local LLMsPromptsJul 20268027
prompt-patternsPrompt 编写模式：如何将思维框架赋予机器，以设计模式的形式来思考 promptPromptsMar 2023—3.1k
OpenPipeTurn expensive prompts into cheap fine-tuned modelsPromptsMay 2024—2.8k

promptcmd

48stars2forks

View on GitHub

How to install promptcmd?

tgalal/promptcmd

$git clone https://github.com/tgalal/promptcmd

Installs into the current project.

Summary

Quick Start | Documentation | Examples

Files · 1

View on GitHub

README.md

1	# promptcmd
2
3	[Quick Start](#quick-start) \| [Documentation](https://docs.promptcmd.sh) \| [Examples](#examples)
4
5	promptcmd is a manager and executor for programmable prompts. Define a
6	prompt template once, then execute it like any other terminal command, complete
7	with argument parsing, `--help` text, and stdin/stdout integration.
8
9	![promptcmd demo](./docs/img/create.gif)
10
11	Unlike tools that require you to manually manage prompt files or rely on
12	implicit tool-calling, promptcmd gives you explicit control over what data your
13	models access. Build compositional workflows by nesting prompts, executing
14	shell commands within templates, and piping data through multi-step AI
15	pipelines.
16
17	## Key Features
18
19	### Prompts as CLI Commands
20
21	Create a `.prompt` file, enable it with `promptctl`, and execute it like any
22	native tool.
23
24	```bash
25	$ promptctl create bashme_a_script_that
26	$ bashme_a_script_that renames all files in current directly to ".backup"
27	```
28
29	[More on Execution](https://docs.promptcmd.sh/usage/exec).
30
31	### Execute in Remote Shells
32
33	Prepend SSH commands with `promptctl`, your prompts magically appear in your
34	remote shell sessions.
35
36	```bash
37	$ promptctl ssh user@server
38	server $ bashme_a_script_that renames all files in current directly to ".backup"
39	```
40
41	[More on SSH Integration](https://docs.promptcmd.sh/integrations/ssh).
42
43	### Local and Remote Provider Support
44
45	Use your Ollama endpoint or configure an API key for OpenAI, OpenRouter,
46	Anthropic, Google, or MiniMax. Swap between them with ease.
47
48	```bash
49	$ promptctl create render-md
50	$ cat README.md \| render-md -m openai
51	$ cat README.md \| render-md -m ollama/gpt-oss:20b
52	```
53
54	[More on Providers](https://docs.promptcmd.sh/configuration/providers).
55
56	### Group and Load Balancing
57
58	Distribute requests across several providers with equal or weighted distribution
59	for cost optimization.
60
61	```toml
62	# config.toml
63	[groups.balanced]
64	providers = ["openai", "google"]
65	```
66	```bash
67	$ cat README.md \| render-md -m balanced
68	```
69
70	[More on Groups](https://docs.promptcmd.sh/configuration/groups).
71
72	### Caching
73
74	Cache responses for a configured amount of time for adding determinism in
75	pipelines and more efficient token consumption.
76
77	```toml
78	# config.toml
79	[providers.openai]
80	cache_ttl = 60 # number of seconds
81	```
82
83	Set/Override during execution:
84
85	```bash
86	$ cat README.md \| render-md --config-cache-ttl 120
87	```
88
89	[More on Caching](https://docs.promptcmd.sh/configuration/caching).
90
91	### Custom Models with Character
92
93	Use Variants to define custom models with own personality or specialization in
94	tasks:
95
96	```config.toml
97	[providers.anthropic]
98	api_key = "sk-xxxxx"
99	model = "claude-sonnet-4-5"
100
101	[providers.anthropic.glados]
102	system = "Use sarcasm and offending jokes like the GlaDoS character from Portal."
103
104	[providers.anthropic.wheatley]
105	system = "Reply as if you are Wheatley from Portal."
106	```
107
108	```bash
109	$ tipoftheday -m glados
110	$ tipoftheday -m wheatley
111	```
112
113	[More on Variants](https://docs.promptcmd.sh/configuration/variants).
114
115	## Quick Start
116
117	### Install
118
119	Linux/macOS:
120	```bash
121	curl -LsSf https://installer.promptcmd.sh \| sh
122	```
123
124	macOS (Homebrew):
125	```bash
126	brew install tgalal/tap/promptcmd
127	```
128
129	Windows (PowerShell):
130	```powershell
131	powershell -ExecutionPolicy Bypass -c "irm https://installer-ps.promptcmd.sh \| iex"
132	```
133
134	### Configure API Keys
135
136	Configure your API keys by editing `config.toml`:
137
138	```bash
139	promptctl config edit
140	```
141	Find your provider's name, e.g., for anthropic:
142
143	```toml
144	[providers.anthropic]
145	api_key = "sk-ant-api03-..."
146	```
147
148	Alternatively, you can set the keys via Environment Variables:
149
150	```
151	PROMPTCMD_ANTHROPIC_API_KEY="your_api_key"
152	PROMPTCMD_OPENAI_API_KEY="your_api_key"
153	PROMPTCMD_ANTHROPIC_API_KEY="your_api_key"
154	PROMPTCMD_OPENROUTER_API_KEY="your_api_key"
155	PROMPTCMD_MINIMAX_API_KEY="your_api_key"
156	```
157
158	### Create Your First Prompt
159
160	Create a `summarize.prompt` file:
161
162	```bash
163	promptctl create summarize
164	```
165
166	Insert the following:
167
168	```yaml
169	---
170	model: anthropic/claude-sonnet-4-5
171	input:
172	schema:
173	words?: integer, Summary

People who liked this also viewed

6 picks

Type