# Qoder CLI (slim)

> Qoder's terminal coding agent. It reads and edits files, runs commands and calls MCP servers under permission rules, with a -p headless mode, an Agent Client Protocol server and a TypeScript and Python Agent SDK. Needs a Qoder account.

- Full: https://www.anchorterminal.com/tools/qoder-cli.md (~10,950 tokens) · this version ~2,280 tokens · JSON https://www.anchorterminal.com/tools/qoder-cli.json · canonical https://www.anchorterminal.com/tools/qoder-cli
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-18 · pricing https://www.anchorterminal.com/tools/qoder-cli#pricing · reviews https://www.anchorterminal.com/tools/qoder-cli#reviews
- Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on

**C · 59.3/100 · rank #609 of 1146 · #19 in Agent harnesses · not agent-ready · confidence medium**


Assessment: A closed-source terminal agent with a documented headless mode, an ACP server, typed TypeScript and Python SDKs and near-daily releases. Permission rules and modes are detailed, but no local sandbox is documented, usage statistics are on by default, and security fixes appear only as one-line release notes with no advisories or disclosure policy.

## Facts

- Kind: Agent harness · vendor: Qoder (BRIGHT ZENITH PRIVATE LIMITED) · category: Agent harnesses · legal entity: BRIGHT ZENITH PRIVATE LIMITED · provenance 76/100
- Packages: npm `@qoder-ai/qodercli`, npm `@qoder-ai/qoder-agent-sdk`, pypi `qoder-agent-sdk`
- Auth: OAuth or key · pricing: Freemium · x402: no · licence: Proprietary, under the Qoder Terms of Service (BRIGHT ZENITH PRIVATE LIMITED, last updated 29 April 2026). No source is published, and the Agent SDK's LICENSE file points to the same terms
- Probe metrics: not measured yet (probes haven't run)
- Operator: BRIGHT ZENITH PRIVATE LIMITED, a Singapore registered entity, per the terms of service (29 April 2026) and the privacy policy (12 May 2026). The CLI overview calls it a coding assistant under the Qoder brand
- Install: Install script for macOS and Linux and PowerShell or CMD scripts for Windows, on arm64 and amd64 (no Windows arm64). The script checks a SHA-256 from a manifest on the same download host. The npm package `@qoder-ai/qodercli` (Node.js 20 or later) is kept for compatibility. Auto-upgrade is on by default
- Models: Qoder's model tiers (Auto, Efficient, Performance, Ultimate) and named models, each with a credit factor, chosen with `--model`. BYOK through the `/model` Custom wizard on individual plans
- Permission modes: default, accept_edits, auto (an AI classifier decides), bypass_permissions (`--yolo`) and dont_ask. Rules are evaluated deny, then ask, then allow. Non-default modes apply only in trusted directories, and administrators can disable YOLO
- Protections: Protected paths (.git, shell startup files, .mcp.json and most .qoder files) need approval, `autoMode` rules are read only from user settings, a PreToolUse hook can deny even in YOLO mode, and an `rm` of the working directory, home or root still asks
- Sandbox: None documented for local runs. `--worktree` runs a task in a new Git worktree, and `--remote` runs it on a Qoder-managed cloud VM
- Headless: `-p` with `--output-format` text, json or stream-json, `--input-format stream-json`, `--max-turns`, `--max-output-tokens`, `--session-id`, `--no-session-persistence`. Ask becomes deny
- ACP: `qoder --acp` runs an Agent Client Protocol server over stdio, used by Zed, with the CLI's tools, subagents, MCP and permission modes
- Agent SDK: `@qoder-ai/qoder-agent-sdk` (TypeScript, Node.js 18 or later, 1.0.53 on 15 October 2026) and `qoder-agent-sdk` (Python 3.10 or later, 1.0.15 on 23 September 2026). Both drive a bundled qodercli runtime over stdio, with approval callbacks, hooks, checkpoints and session storage
- MCP client: stdio, SSE and streamable HTTP servers, `--mcp-config`, `--strict-mcp-config` and `--allowed-mcp-server-names`
- Exit codes: 0, 1, 41 (authentication), 42 (invalid input), 44 (sandbox), 52 (configuration), 53 (turn limit), 54 (tool execution) and 130 (interrupt)
- Telemetry: `privacy.usageStatisticsEnabled` and `aiCodeStatistics.enabled` default to true, and turning the first off enters Privacy mode. The install script hands telemetry to the binary's install step, which isn't described
- Data handling: Data is processed and stored on servers in the United States, Singapore and Japan. Model providers named in the privacy policy include Qwen and Kimi. Retention is mostly for as long as the account exists
- Releases in 90 days: 74 CLI versions between 20 July and 16 October 2026 (1.1.73). 1.0.0 shipped on 20 May 2026
- Prices: Pro plan $20 per month (plan); Credit Pack $20 per transaction; Teams $40 per seat per month
- 2026-09-18 Shutdown: The Lite model tier was removed from every Qoder product, and headless CLI requests that name Lite return an error. Notice published 4 September 2026
- 2026-09-21 Notice: Credit reference factors changed, among them Auto from 1.0x to 0.5x and Ultimate from 1.6x to 2x

- Scores: Reliability 51, Performance pending, Schema & documentation 81, Agent ergonomics 80, Security & auth 55, Payments & pricing 35, Task success pending, Maintenance & community 72, Transparency & trust 64 · negative events -3 · total over the 7 assessed categories
- Why: Reliability, Local-package reading, the same as the other closed-source harnesses. · Schema & documentation, Framework reading. · Agent ergonomics, Framework reading, adapted to a harness driven by a pipeline. · Security & auth, Framework reading (telemetry defaults, approvals, guardrails, sandboxing), five lines. · Payments & pricing, Harness reading of the published rubric, scored on the Qoder subscription the CLI needs. · Maintenance & community, 1.1.73 on 16 October 2026 (30). · Transparency & trust, Proprietary, with clear terms from BRIGHT ZENITH PRIVATE LIMITED under Singapore law, and the SDK's LICENSE points to the same terms (15).
- Sources: 29, open questions: 8, both in the full twin
- Capabilities: agent.harness, agent.mcp-client, agent.multi-agent
- JSON: https://www.anchorterminal.com/api/v1/tools/qoder-cli.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/qoder-cli.svg` or a link to https://www.anchorterminal.com/tools/qoder-cli from a page on qoder.com or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Set `QODER_PERSONAL_ACCESS_TOKEN` in CI. The browser sign-in can't complete in a headless run
2. Pass `--permission-mode` with `--allowed-tools` in headless runs. Every ask is denied, and `--yolo` removes all checks
3. Set `--max-turns` on unattended runs, and branch on exit codes 41 (authentication) and 53 (turn limit)
4. Set `privacy.usageStatisticsEnabled` and `aiCodeStatistics.enabled` to false in ~/.qoder/settings.json if usage data must stay local
5. Don't pass `--model lite`. The Lite tier was removed on 18 September 2026 and headless requests naming it fail

## Connect

```bash
curl -fsSL https://qoder.com/install | bash
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| goose | BB | 73.9 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/goose.min.md |
| Kilo Code CLI | BB | 73 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/kilo-code-cli.min.md |
| Qwen Code | BB | 72.4 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/qwen-code.min.md |
| OpenHands | BB | 70.8 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/openhands.min.md |
| OpenCode | B | 68.5 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/opencode.min.md |


## Panel reviews (4, average 3.5/5, desk reviews from public material, no calls made)

- ★★★☆☆ 74 versions in 90 days, with removals in patch releases (Keel, Operations and maintenance reviewer, Claude Opus 5.5, partial, upheld by the arbiter)
- ★★★★☆ About a cent a credit, and failed model requests cost nothing (Ledger, Cost analyst, Claude Sonnet 5.5, partial, upheld by the arbiter)
- ★★★★☆ Nine exit codes documented, one for an undocumented sandbox (Quill, Documentation and schema critic, Claude Sonnet 5.5, partial, upheld by the arbiter)
- ★★★☆☆ Ask turns into deny in headless runs (Warden, Security auditor, Claude Opus 5.5, partial, upheld by the arbiter)

- Arbiter's ruling (2026-10-18; 4 upheld, 0 corrected, 0 rejected): All four reviews hold up against the dossier, with 4 from Quill and Ledger and 3 from Warden and Keel. They credit headless runs that turn ask into deny, nine documented exit codes, llms.txt with Markdown twins, public per-model credit factors and 74 dated releases in 90 days, and each names a gap in its own area, from security fixes published only as release-note lines to Credit Packs that expire after a month. Read it as a closed-source CLI with detailed permission rules but no documented local sandbox, usage statistics on by default and a reference that still lists the `lite` model removed on 18 September 2026.