EntelligenceAI CLI
AI-powered code review assistant that helps you catch bugs, improve code quality, and follow best practices - all from your terminal.
🚀 Installation
From PyPI
pip install entelligence-cli
From Source (this repo)
# Clone the repo and install in editable mode
git clone https://github.com/Entelligence-AI/cli.git
cd cli
uv tool install --editable .
If uv is not installed:
curl -LsSf https://astral.sh/uv/install.sh | sh
Uninstall
uv tool uninstall entelligence-cli
Auto-update
The CLI keeps itself up to date automatically. On any command it checks PyPI
(at most once per day) and, when a newer version exists, upgrades in place using
your install method (uv tool / pipx / pip) and re-runs your command on the
new version. Auto-update is skipped for non-interactive/piped output, CI, and
editable/dev installs, where it instead prints a one-line "update available" notice.
To opt out (e.g. pinned or centrally managed installs), set the env var or config flag:
export ENTELLIGENCE_NO_AUTO_UPDATE=1
# or, in ~/.entelligence/config.json: { "auto_update": false }
Enterprise / managed installs
Releases are published from CI via PyPI Trusted Publishing (OIDC — no long-lived API token) with PEP 740 provenance attestations, so every version on PyPI is verifiable back to this repository's publish workflow.
For a centrally-managed fleet, install a pinned version and turn off unattended self-update, so a new release rolls out only when you choose:
# Pin the version the installer lays down (a known-good release). The assignment goes on the
# piped `sh` so the installer shell inherits it — `VAR=x curl ... | sh` would scope VAR to
# curl only, and the installer would still float to the latest release:
curl -fsSL https://app.entelligence.ai/install.sh | ENTELLIGENCE_CLI_VERSION=0.1.246 sh
# (equivalent: export it first, then run the installer)
export ENTELLIGENCE_CLI_VERSION=0.1.246
curl -fsSL https://app.entelligence.ai/install.sh | sh
# Disable unattended auto-update — the fleet controls the upgrade cadence:
export ENTELLIGENCE_NO_AUTO_UPDATE=1
📝 Quick Start
1. Get Your API Key
Sign up and get your API key: app.entelligence.ai/settings?tab=api
2. Authenticate
entelligence auth login
Paste your API key when prompted.
3. Review Your Code
entelligence review
That's it! The CLI will analyze your uncommitted changes (or committed changes if none) and provide intelligent feedback.
📖 Usage
Authentication Commands
# Log in with your API key
entelligence auth login
# Check authentication status
entelligence auth status
# Log out
entelligence auth logout
Review Commands
# Review current changes (default: uncommitted, falls back to committed)
entelligence review
# Review only committed changes vs base branch
entelligence review --committed-only
# Review against a specific branch
entelligence review --base-branch develop
# Set review priority
entelligence review --priority high
# Verbose output mode
entelligence review --mode verbose
# Plain text output
entelligence review --plain
# Debug mode
entelligence review --debug
Insights Commands
Coding-agent analytics powered by cinsights.
Every cinsights <cmd> is available as entelligence insights <cmd>. It reads
your local Claude Code / Codex session files, scores them, and (optionally)
runs LLM analysis to surface friction patterns, per-project digests, and
per-developer profiles.
Onboarding is two commands. sync sets up both lanes: it processes your
sessions locally and pushes the computed session + task metrics (the breadth
lane, which every dashboard number is built from), and it installs the
background activity monitor so collection continues even when a session stays
open for days. The monitor checks and delivers pending session changes on startup
and then hourly, including sessions that remain continuously active.
Failed or busy monitor deliveries retry on the next hourly attempt without needing new turns.
The independent metrics job checks every 15 minutes. Its 15-minute completion
throttle can skip one tick after a run, leaving processing headroom compared
with the previous hourly trigger and one-hour throttle. Existing jobs adopt
the new interval when a successful sync re-arms them; fixed-time schedules stay fixed.
Upload and backend processing time are additional.
Existing delivery checkpoints remain authoritative across restarts.
Sessions that parse but have no analyzable turns remain incomplete. The analysis
worker defers unchanged versions of these sessions before applying its run limit,
so they cannot repeatedly displace healthy work. They retry when their local input
changes, the CLI version changes, a new replay directive arrives, or analysis is
forced. This local hint never acknowledges a replay or supplies missing billing.
On a supporting backend, the inventory cycle also starts an independent daily billing worker, even when no sessions need processing. It collects exact completed UTC days of Cursor account usage and Claude reference usage from local transcript requests. Claude amounts are rate-card references, not Anthropic invoices, and cloud-only history absent from local homes is not covered. Billing errors or hung collectors never hold up regular session sync; incomplete days remain retryable. Large days use acknowledged chunks before publication. See daily billing behavior, coverage, and limits.
The raw (redacted) transcript upload — the depth lane for decisions, replay and
byte-derived pricing — runs alongside metrics syncs only for orgs whose
backend posture permits it. Unchanged transcripts are skipped. The 15-minute metrics
and daily raw jobs remain recovery backstops; an hourly supervisor restarts the
activity monitor if it exits. --no-schedule keeps a manual sync one-off.
entelligence reset --schedule stops the monitor and removes its supervisor.
entelligence auth login
entelligence insights sync
Narrowing flags if you need them: --transcripts runs the transcript upload
on its own (this is what the scheduled depth-lane job invokes); --metrics-only
is the explicit spelling of what a plain sync already does, kept for scripts
that state it. Everything is redacted client-side before upload.
# One-time: configure the LLM provider (saved to ~/.cinsights/config.json)
entelligence insights setup
# Index local sessions (free, no LLM calls)
entelligence insights index --source local --hours 8760
# Analyze sessions (shows a cost estimate and asks for confirmation)
entelligence insights analyze --source local
# Generate a project or developer digest
entelligence insights digest project <my-project> --days 30
entelligence insights digest user <user-id> --days 30
# Launch the local web dashboard (http://localhost:8100)
entelligence insights serve
# Full subcommand list and flags
entelligence insights --help
Common Workflows
Before Committing:
git add .
entelligence review
git commit -m "Your message"
Before Creating PR:
git checkout -b feature/my-feature
# ... make changes ...
entelligence review --priority high
git push origin feature/my-feature
🔧 Configuration
Config File
Configuration is stored at ~/.entelligence/config.json with secure permissions (read/write for owner only).
Environment Variables
For non-interactive use, you can set:
ENTELLIGENCE_API_KEY- Your API key (alternative toentelligence auth login)
💡 Benefits
For Individual Developers
- Catch bugs before they reach code review
- Learn best practices through AI suggestions
- Save time on code reviews
- Improve code quality consistently
For Teams
- Maintain consistent code standards
- Reduce code review time
- Improve code quality across the team
🆘 Troubleshooting
Authentication Failed
# Check your authentication status
entelligence auth status
# If invalid, re-authenticate
entelligence auth logout
entelligence auth login
Ensure your API key is valid at app.entelligence.ai/settings?tab=api
Connection Timeout
- Check your internet connection
- Try again with
--debugflag for more information
No Changes Detected
# Check what changes are available
git status
# Review only committed changes if you have commits on your branch
entelligence review --committed-only
# Or review against a different base branch
entelligence review --base-branch develop
📚 Resources
- Website: entelligence.ai
- Documentation: docs.entelligence.ai
- Support: info@entelligence.ai
📄 License
Proprietary - Copyright © 2026 EntelligenceAI. All rights reserved.
❓ FAQ
Q: Is my code stored on your servers? A: No, code is only analyzed transiently and not permanently stored.
Q: Does this work with private repositories? A: Yes, all code remains private and secure.
Q: What languages are supported? A: Python, JavaScript, TypeScript, Java, Go, Rust, C++, Ruby, PHP, Swift, Kotlin, and more.
Q: Can I review specific files only? A: Currently reviews are based on Git changes. File-specific reviews coming soon.
Made with ❤️ by EntelligenceAI
1 maintainer
Release files for entelligence-cli 0.1.350
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| entelligence_cli-0.1.350.tar.gz | 1.5 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| entelligence_cli-0.1.350-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.6 MB
Release files / entelligence_cli-0.1.350.tar.gz
| Download URL | entelligence_cli-0.1.350.tar.gz |
|---|---|
| Size | 1.5 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1f31505e402ff91c121962ea61f9a03ca5989da9676bb3a1e7b267aa30d2674f
|
|
BLAKE2b-256 checksum How to use checksums |
37d0659c2d03bc99c9489777fd52ffbc81ab00397b48e38a88f9208eaef65d00
|
| 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 27, 2026.
Transparency logRelease files / entelligence_cli-0.1.350-py3-none-any.whl
| Download URL | entelligence_cli-0.1.350-py3-none-any.whl |
|---|---|
| Size | 1.1 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a281be20286791abecce1d8e68556c782172915356b63228e619e66af76901e6
|
|
BLAKE2b-256 checksum How to use checksums |
e1997eb0f7c871a034330a2528908f6a8ee98c2ceaf110cdbb932e9d42669482
|
| 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 27, 2026.
Transparency log
