CLI Agent
🤖 AI Drafted (Not reviewed)
Run a local agent CLI (Claude/Codex) and capture stdout/stderr.
|
|
| Tool id |
cli_agent |
| Category |
agent |
| Uses a language model |
no |
| Needs a generative model |
no |
| Runs over many items |
no |
| Item handling |
batch |
| Structured output |
no |
| Human-verified |
not yet |
What it reads
| Port |
Type |
Required |
What it is |
Task (task) |
text |
yes |
Prompt/task for the CLI agent. |
Context (context) |
any |
no |
Optional structured context included in prompt template. |
What it emits
| Port |
Type |
Required |
What it is |
Text (text) |
text |
— |
— |
Stdout (stdout) |
text |
— |
— |
Stderr (stderr) |
text |
— |
— |
Exit Code (exit_code) |
number |
— |
— |
Options
| Option |
Type |
Default |
What it does |
cli |
string |
claude |
CLI to invoke. One of: claude, codex. |
ollama_model |
string |
qwen2.5-coder |
Model to use when use_ollama_launch=true. |
prompt_template |
string |
— |
Template using {task}, {context}, {cli}. |
timeout_seconds |
integer |
120 |
Execution timeout in seconds. |
use_ollama_launch |
boolean |
no |
Route execution via ‘ollama run ‘. |
The prompt it sends
This tool does not send a prompt — it does its work in code.