Qoder CLI is the terminal coding agent of the Qoder platform: it analyses codebases, edits files, runs commands and connects MCP servers to turn a language model into a working AI harness for developers in the terminal. According to its terms of service, the platform is operated by Bright Zenith Private Limited. As of 20 September 2026.
What is Qoder CLI?
According to its release notes, Qoder CLI has been publicly available since version 0.1.0 on 15 October 2025 and runs on macOS, Linux and Windows (including WSL). According to the terms of service and privacy policy, the platform’s operator is Bright Zenith Private Limited, a Singapore-registered company. The software is proprietary, and no public source-code repository is documented. Qoder CLI is part of a wider product line that also includes an IDE, a JetBrains plugin and cloud agents; the CLI itself is aimed, per the vendor, at developers who want to work directly in the terminal without switching to a graphical interface.
Qoder CLI as an AI harness
Qoder CLI offers several operating modes: an interactive default mode, a Plan mode for analysis before changes, a Goal mode for continuous goal pursuit, a Scheduled Task mode, /loop for recurring tasks, and a headless mode for scripts and CI/CD pipelines. Project instructions live in AGENTS.md at the project root or under ~/.qoder/AGENTS.md for account-wide rules, supplemented by project-specific rules under .qoder/rules/**/*.md. Permissions are governed by a six-tier system (default, accept_edits, auto, bypass_permissions, dont_ask, plan) with fine-grained allow/ask/deny rules such as Bash(npm run test:*) in settings.json, merged across several levels – user, project and local settings, plus CLI arguments. Via MCP, external tools can be connected with qoder mcp add <name> -- <server command>, and MCP calls remain subject to the same permission rules. For spec-driven development, Quest mode, per the vendor, splits the work into a Design Agent that produces a specification and a Task Executor that implements it.
Models and model access
Qoder CLI provides a curated model catalogue that, according to the documentation, changes over time and by account; the vendor does not specify a fixed, publicly named list of supported model families. According to the privacy policy, Alibaba’s Qwen and MoonShot’s Kimi models are among the language models used in the coding experience. In addition, a BYOK wizard (/model → Custom) lets you add your own model credentials – the documentation explicitly advises against manually configuring this in settings.json, since the available providers and fields are determined by the account’s current BYOK catalogue. Whether model access can also be routed through a freely configurable custom endpoint or a central proxy is not publicly documented. Organisations that still want to centrally govern and budget model access can place innFactory’s AI Gateway in front of the model APIs they already use, as an OpenAI-compatible control layer.
Enterprise use: data protection and governance
According to the privacy policy, the vendor processes inputs and the outputs generated from them in locations including the United States, Singapore and Japan, and uses them for service improvement depending on the plan; a “Share & Improve” setting allows an opt-out depending on the plan, per the vendor. The public policy names no dedicated EU or GDPR addendum, nor does it mention Zero Data Retention or a separate data processing agreement for enterprise customers. Important in practice: source code and prompts are sent to the chosen model or service provider – this data flow belongs in an organisation’s records of processing activities in advance, together with a check of current terms directly with the vendor.
Orchestrating Qoder CLI with herdr
According to the documentation of the terminal orchestrator herdr, it detects Qoder CLI sessions via a screen manifest and integrates them at the session level. This makes it possible to monitor and control Qoder CLI alongside other coding agents from one shared terminal interface.
Where it fits: developer harness and enterprise harness
Qoder CLI is a harness built for one developer working in one terminal. For business users outside development, CompanyGPT is the matching harness: tools via MCP servers and n8n workflows, context via companyRAG, roles from Entra ID or Keycloak, and audit logs for business departments. The AI Gateway forms the shared cost and governance layer for both worlds. For an overview of other coding agents and orchestrators, see our overview of all AI harnesses. We help organisations adopt coding harnesses such as Qoder CLI securely, with proper governance and centralised cost control.