ai-flow-cli
Lightweight, fast terminal AI wrapper around modern coding harnesses with Rich markdown and math rendering.
Wraps fast headless AI harnesses (pi, omp, claude, codex, copilot, opencode) for everyday queries and Unix pipelines.
⚡ Features
- 🚀 Multi-Harness Support — seamlessly autodetect and use
pi,omp,claude,codex,copilot, oropencode. - 🧙 Interactive Setup Wizard — run
ai --wizardto inspect detected harnesses and choose your default. - 🎨 Rich Terminal Markdown & LaTeX — renders clean markdown, tables, syntax highlighting, and math formulas in the terminal.
- 💬 No Quotes Required — run
ai what is the biggest object on earthdirectly. - 🚰 Pure Unix Pipelines — stdin context injection and clean raw text output when piped.
- 🛠️ Tools On Demand — pass
--toolswhen agentic filesystem/system actions are needed.
📦 Installation
1. Arch Linux (AUR)
yay -S ai-flow-cli
# or paru:
paru -S ai-flow-cli
2. Recommended: Install via mise
mise use -g pipx:ai-flow-cli
# or via python backend:
mise use -g pip:ai-flow-cli
3. Install via PyPI (pip / pipx / uv)
# Standard pip install:
pip install --user ai-flow-cli
# Isolated global CLI with pipx:
pipx install ai-flow-cli
# Fast install with uv:
uv tool install ai-flow-cli
4. One-Line Script Install
curl -sSL https://raw.githubusercontent.com/Codder13/ai-flow-cli/main/install.sh | bash
🛠️ Usage
1. Initial Setup & Harness Selection Wizard
Run the interactive wizard to detect installed harnesses and set your preferred default:
ai --wizard
If no config exists yet, ai automatically prompts the wizard on first run when interactive, or picks the first detected harness (pi -> omp -> claude -> ...).
2. Direct Questions (Fast Headless Mode)
No quotes needed:
ai explain what is eBPF in 3 bullet points
3. Switching Agent Harnesses on the Fly
Use -a or --agent (or --harness) to override your default agent harness for a single query:
ai -a claude explain why rust ownership works this way
ai -a omp write a bash script to backup my dotfiles
4. Handoff to Harness TUI (-H / --handoff)
When a terminal query evolves into a deeper interactive agent session, hand off your entire conversation context to a full harness TUI:
# Handoff current context to default or specified harness TUI
ai -H
ai -H omp
ai -H claude "Fix all broken unit tests across the whole workspace"
5. Enable Tools for Agentic Actions
By default, queries run without tool side-effects for speed and safety. Pass --tools or -t to enable filesystem and command execution:
ai --tools check git status and run tests
6. Unix Pipelines & Input Redirection
When piped, ai preserves standard Unix conventions:
git diff | ai explain these changes
cat error.log | ai "what caused this panic?"
curl -s https://example.com | ai summarize this page
Output is automatically raw and unformatted when redirected to a pipe or file:
ai "generate a python regex for uuid4" > regex.txt
⚙️ Configuration
Settings are stored in ~/.config/ai/config.json:
{
"harness": "pi"
}
You can change it anytime via:
ai --wizard
📄 License
MIT
Metadata
Release files for ai-flow-cli 0.10.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| ai_flow_cli-0.10.0.tar.gz | 23.8 kB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ai_flow_cli-0.10.0-py2.py3-none-any.whl | Python 2, Python 3 | none | any | Details |
Total release size: 48.8 kB
Release files / ai_flow_cli-0.10.0.tar.gz
| Download URL | ai_flow_cli-0.10.0.tar.gz |
|---|---|
| Size | 23.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
badacd0786791757d0be26fe3386f917a98058c68d0ada0e81cf890ff6ad3150
|
|
BLAKE2b-256 checksum How to use checksums |
a7cfe008007e128ce5951db4203c54058f2454a61c4ab7e9932f00391b7816b6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 21, 2026.
Transparency logRelease files / ai_flow_cli-0.10.0-py2.py3-none-any.whl
| Download URL | ai_flow_cli-0.10.0-py2.py3-none-any.whl |
|---|---|
| Size | 25.1 kB |
| Tags | Python 2 Python 3 |
|
SHA-256 checksum How to use checksums |
793ecb7b8c140037d892ecf7f389599ea0946eb62c5059c239dc4b19e54051c7
|
|
BLAKE2b-256 checksum How to use checksums |
fd3caacd9e4465840802ecebd7cfdb61013eb0acc9a1cc4d37ae65645c26fcb3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 21, 2026.
Transparency logSecurity for ai-flow-cli 0.10.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for ai_flow_cli-0.10.0.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 21, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.