> ## Documentation Index
> Fetch the complete documentation index at: https://tashfiq.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.
# Quickstart
> Get a real Charter score on your repo in under 5 minutes.
This guide takes you from zero to a real Charter scan result. By the end you'll have a score, a list of findings to act on, and an offline report you can share.
The fastest path on macOS and Linux:
```bash theme={null}
brew install use-charter/tap/charter
```
Verify the install:
```bash theme={null}
charter version
```
On Windows or if you prefer a direct binary or `go install`, see the [Installation guide](/docs/installation) for all supported platforms.
From the root of the repository you want to evaluate:
```bash theme={null}
charter doctor --path .
```
Charter scans statically — no network requests, no LLM calls, nothing sent anywhere. On a typical repo it finishes in under 2 seconds.
The doctor output has three sections:
**1. Category scorecard**
A breakdown of findings by axis — Context, Safety, Operability, and Governance — so you can see which area is pulling the score down.
**2. Individual findings**
Each finding shows:
* Rule ID (e.g. `AE-SEC-001`)
* Severity (Blocker / High / Medium / Low)
* Evidence — the exact file or pattern that triggered the rule
* A fix suggestion or a link to the remediation guide
**3. Final score and gate result**
The total score (0–100) and whether the repo passed or failed the active threshold.
**Exit code contract:**
* Exit `0` — passed threshold
* Exit `1` — below threshold
* Exit `2` — scan error (bad path, unreadable files, etc.)
Use the score as a quick triage signal:
| Score | Zone |
| ------ | -------------------------------- |
| 80–100 | Ship-ready |
| 60–79 | Needs work |
| 50–59 | Blocked (active Blocker finding) |
| 0–49 | Critical (raw secret detected) |
For findings Charter knows how to repair, preview the diff first:
```bash theme={null}
charter fix --dry-run
```
Apply when you're satisfied:
```bash theme={null}
charter fix
```
Secrets and dangerous-command findings are never auto-fixed. Charter shows you what needs to change and where, but you make those changes manually. This is intentional — automated secret handling requires decisions only a human should make.
After fixing, re-run `charter doctor --path .` to confirm the score improved.
For a fuller offline review you can attach to a PR or hand to a colleague:
```bash theme={null}
charter report
```
This writes `charter-report.html` — a self-contained file with no external network dependencies. Open it in any browser.
## What's next
Gate pull requests on agent-readiness with the GitHub Action.
The scoring formula, hard caps, and how findings roll up.
All 18 rules with remediation guidance for each.
When auto-fix isn't available, here's how to close findings by hand.