C

Cline

by Cline

Developer ToolsAI Agents & OrchestrationAgent Development

Open-source coding agent that plans before it acts, in your IDE or your terminal

Freemium · Usage-based · Subscription · Contact for pricing·Added Mar 11, 2026·Updated Aug 23, 2026
Share:
THE DAILY BRIEF
Cline

by Cline

Developer ToolsAI Agents & OrchestrationAgent Development

Open-source coding agent that plans before it acts, in your IDE or your terminal

Freemium · Usage-based · Subscription · Contact for pricing

Cline is an open-source coding agent for VS Code, JetBrains and the terminal that reads a repository, edits files, runs commands and drives a browser under human approval. It is aimed at engineering teams that want agentic coding without vendor lock-in — bringing their own model keys and running the agent inside their own VPC, on-premises or air-gapped network.

At a Glance

Category
Developer Tools
Pricing
Freemium, Usage-based, Subscription, Contact for pricing
Target Market
CTOs, VPs of Engineering, Platform Engineers, Enterprise Developers, Security-Constrained Engineering Teams
Deployment
Open-source, Self-hosted, Hybrid, Cloud-first
Customers
Roughly 5 million installs across the VS Code Marketplace and Open VSX as of January 2026; Samsung, SAP and other Fortune 100 engineering teams named publicly

Key Features

  • ✓Plan/Act mode
  • ✓Checkpoints
  • ✓No codebase indexing
  • ✓Bring-your-own-inference
  • ✓MCP marketplace
  • ✓Headless CLI and SDK
  • ✓Self-hosted deployment

Capabilities

✓text generation
✗image generation
✗video generation
✓code generation
✓workflow automation
✓api access
✗audio generation
✗fine tuning
✓agent orchestration

Use Cases

  • •Legacy modernisation
  • •Vulnerability remediation
  • •Test generation for untested code
  • •Automated code review in CI
  • •Agentic coding under data-residency constraints

Ideal For

Best For

  • ✓Enterprises that must keep source code inside their own network and route inference through their existing Bedrock, Vertex or Azure OpenAI contracts
  • ✓Teams already paying for model access who want to avoid a per-seat coding-assistant subscription layered on top of token costs
  • ✓Legacy modernisation, vulnerability remediation and test generation work where a plan can be reviewed before any file is touched
  • ✓Platform teams embedding a coding agent into CI/CD via the headless CLI, for tasks like automated code review or documentation upkeep
  • ✓Developers running local models through Ollama or LM Studio who need an agent that does not assume a hosted backend

Not Ideal For

  • ✗Teams that need a predictable flat monthly bill — bring-your-own-key means costs scale with usage, and heavy sessions with frontier models are widely reported to run several dollars per hour
  • ✗Very large monorepos where semantic search over an index genuinely helps; Cline deliberately does not index the codebase, and HN commenters running tens of thousands of reference files consider that disqualifying
  • ✗Organisations without the security maturity to vet an agent that edits files and executes shell commands — the February 2026 npm supply-chain compromise showed the blast radius
  • ✗Non-developers or business users looking for a no-code app builder; Cline assumes a repository, a terminal and someone who can review a diff

Market Analysis

Open-sourceDeveloper-firstEnterprise-ready

Pros

  • ✓Genuinely open source under Apache 2.0, so the agent's prompting and tool-use logic can be read, forked and audited — several HN commenters said they learned agent design from the codebase
  • ✓No vendor lock-in on models: the same agent runs against frontier APIs, your own Bedrock or Vertex contract, or a local model through Ollama
  • ✓Enterprise deployment options run to VPC, on-premises and air-gapped, which is rare among coding agents and decisive for regulated buyers
  • ✓Strong practitioner reputation for long autonomous runs — reports of planning for ten to fifteen minutes then producing an accurate multi-file diff unattended

Cons

  • ✗Token-hungry by design: it sends extensive context with each request, and independent 2026 comparisons put active frontier-model sessions at roughly $5-15 per hour, which is unpredictable against a flat $20/month competitor
  • ✗No codebase indexing is a real limitation on very large repositories, and HN critics maintaining tens of thousands of reference files call semantic search a killer feature Cline is missing
  • ✗Multiple practitioners on Hacker News report Claude Code reaching a correct result in fewer prompts on complex tasks
  • ✗A compromised npm publish token let an attacker ship a malicious Cline CLI 2.3.0 for about eight hours on 17 February 2026, downloaded roughly 4,000 times; the root cause was prompt injection into the project's own AI-powered issue-triage workflow, since removed in favour of OIDC provenance publishing
  • ✗Accuracy is reported to degrade once context grows past a few tens of thousands of tokens, so very large tasks still need decomposition

Pricing

Open Source

$0

  • ✓Full VS Code extension and CLI
  • ✓Bring your own API keys
  • ✓MCP marketplace access
  • ✓Client-side architecture
  • ✓Pay only for model inference

Teams

From $20/user/mo

  • ✓Centralised billing and usage tracking
  • ✓Team management dashboard
  • ✓Budgets and access controls
  • ✓First seats free per Sacra's reporting

Enterprise

Contact for pricing

  • ✓SSO and OIDC
  • ✓Role-based access control and audit logs
  • ✓Global model and MCP policy enforcement
  • ✓VPC, on-premises and air-gapped deployment
  • ✓JetBrains extension
  • ✓SLA and dedicated support

The agent itself is free and Apache-2.0 licensed; the real cost is model inference, which you pay either directly through your own provider keys or through Cline's provider at cost with no markup — an explicit contrast with coding tools that embed margin in token usage. That makes it cheaper than a flat subscription for light users and more expensive for heavy ones: independent 2026 comparisons put active sessions with frontier models at roughly $5-15 per hour and typical monthly API spend between $25 and $200. Teams is reported at $20 per user per month, and Enterprise pricing for SSO, audit logging and self-hosted deployment is not published.

Security & Compliance

✗soc2
✗gdpr
✗hipaa
✗iso27001
✓sso
✓data residency

THE DAILY BRIEF

Enterprise AI insights for technology and business leaders, weekly.

beri.net

Subscribe at beri.net/subscribe for weekly AI insights delivered to your inbox.

LinkedIn: linkedin.com/in/rberi  |  X: x.com/rajeshberi

© 2026 Rajesh Beri. All rights reserved.

Cline is an open-source coding agent for VS Code, JetBrains and the terminal that reads a repository, edits files, runs commands and drives a browser under human approval. It is aimed at engineering teams that want agentic coding without vendor lock-in — bringing their own model keys and running the agent inside their own VPC, on-premises or air-gapped network.

Cline is an open-source coding agent that runs inside VS Code and JetBrains IDEs, as a headless CLI and as an embeddable SDK, editing files, executing terminal commands and driving a browser under the developer's supervision rather than autocompleting single lines. Its defining design choice is Plan/Act mode: in Plan mode the agent explores the codebase, asks clarifying questions and lays out a strategy, and only once the developer agrees does it switch to Act mode and execute, with every change tracked by checkpoints that can be rolled back. Equally deliberate is what it refuses to do — Cline does not index or embed the codebase, arguing that on-demand file traversal beats vector retrieval for correctness. The May 2025 post defending that position drew 178 points and 124 comments on Hacker News, where practitioners split sharply: several argued Cursor's index is a real advantage on very large repositories and called the stance damage control, while others reported leaving Cline to work for ten to fifteen minutes and returning to a diff that was 'completely on the money'. The project is Apache-2.0 licensed with roughly 66,700 GitHub stars and works with Anthropic, OpenAI, Google, OpenRouter, AWS Bedrock, Azure and GCP Vertex, Cerebras, Groq, Ollama, LM Studio and any OpenAI-compatible endpoint, consuming Model Context Protocol servers through a built-in marketplace. Founder Saoud Rizwan raised $32M in total, including a $27M Series A in July 2025 led by Emergence Capital at a $110M post-money valuation with Pace Capital, 1984 Ventures, Essence VC and Cox Exponential participating, on the back of installs that grew from one million in March 2025 to 2.7 million that July and roughly five million by January 2026. Commercially it is bottom-up: free and bring-your-own-key for individuals, Teams at $20 per user per month, and an Enterprise tier adding SSO, audit logs, centrally enforced model and MCP policy, and VPC, on-premises or air-gapped deployment. Samsung and SAP are publicly named users.

Ideal Buyer

A platform or developer-productivity lead in a security-constrained enterprise who wants agentic coding but cannot send source code to a third-party SaaS or accept a vendor's model markup — Cline runs in your VPC against your own Bedrock, Vertex or Azure OpenAI endpoint.

Key Benefit

Developers get a full agentic coding loop — plan, edit, run, test, roll back — with the code, the inference endpoint and the audit trail all staying inside infrastructure the organisation controls.

At a Glance

Category
Developer Tools
Pricing
Freemium, Usage-based, Subscription, Contact for pricing
Target Market
CTOs, VPs of Engineering, Platform Engineers, Enterprise Developers, Security-Constrained Engineering Teams
Deployment
Open-source, Self-hosted, Hybrid, Cloud-first
Customers
Roughly 5 million installs across the VS Code Marketplace and Open VSX as of January 2026; Samsung, SAP and other Fortune 100 engineering teams named publicly

Key Features

  • ✓
    Plan/Act mode

    The agent explores the codebase and proposes a strategy first, and only executes after the developer approves it, which keeps large refactors reviewable.

  • ✓
    Checkpoints

    Every change the agent makes is snapshotted so a bad multi-file edit can be reversed in one action rather than untangled by hand.

  • ✓
    No codebase indexing

    Cline reads files on demand instead of building a vector index, a deliberate correctness trade-off that also means no embedding of proprietary source into a third-party store.

  • ✓
    Bring-your-own-inference

    Connect Anthropic, OpenAI, Google, Bedrock, Vertex, Azure OpenAI, Groq, Cerebras, Ollama or any OpenAI-compatible endpoint, with tokens passed through at cost and no markup.

  • ✓
    MCP marketplace

    Model Context Protocol servers extend the agent to internal systems, and Enterprise administrators can centrally control which servers and models a team may use.

  • ✓
    Headless CLI and SDK

    The same agent runs unattended in CI/CD pipelines or embedded in another product, not just interactively inside an editor.

  • ✓
    Self-hosted deployment

    Enterprise supports VPC, on-premises and air-gapped installation, so regulated organisations keep both the code and the audit trail internal.

Capabilities

✓text generation
✗image generation
✗video generation
✓code generation
✓workflow automation
✓api access
✗audio generation
✗fine tuning
✓agent orchestration

Use Cases

  • •
    Legacy modernisation

    Point the agent at an aging service, review the migration plan it produces, then let it execute file by file with checkpoints to roll back.

  • •
    Vulnerability remediation

    Feed scanner findings to the agent so it patches each affected call site consistently across a repository instead of one ticket at a time.

  • •
    Test generation for untested code

    The agent reads existing modules, writes tests, runs them in the terminal and iterates until the suite passes without a developer babysitting it.

  • •
    Automated code review in CI

    The headless CLI runs against a pull request in the pipeline and posts findings, giving smaller teams a consistent first-pass reviewer.

  • •
    Agentic coding under data-residency constraints

    A regulated enterprise runs the agent air-gapped against a self-hosted model so no source code leaves its own network boundary.

Ideal For

Best For

  • ✓Enterprises that must keep source code inside their own network and route inference through their existing Bedrock, Vertex or Azure OpenAI contracts
  • ✓Teams already paying for model access who want to avoid a per-seat coding-assistant subscription layered on top of token costs
  • ✓Legacy modernisation, vulnerability remediation and test generation work where a plan can be reviewed before any file is touched
  • ✓Platform teams embedding a coding agent into CI/CD via the headless CLI, for tasks like automated code review or documentation upkeep
  • ✓Developers running local models through Ollama or LM Studio who need an agent that does not assume a hosted backend

Not Ideal For

  • ✗Teams that need a predictable flat monthly bill — bring-your-own-key means costs scale with usage, and heavy sessions with frontier models are widely reported to run several dollars per hour
  • ✗Very large monorepos where semantic search over an index genuinely helps; Cline deliberately does not index the codebase, and HN commenters running tens of thousands of reference files consider that disqualifying
  • ✗Organisations without the security maturity to vet an agent that edits files and executes shell commands — the February 2026 npm supply-chain compromise showed the blast radius
  • ✗Non-developers or business users looking for a no-code app builder; Cline assumes a repository, a terminal and someone who can review a diff

Integrations

✓SDK Available
SDK:TypeScript

Deployment

✓On-Premise

Market & Ratings

Estimated Customers

Roughly 5 million installs across the VS Code Marketplace and Open VSX as of January 2026; Samsung, SAP and other Fortune 100 engineering teams named publicly

Market Analysis

Open-sourceDeveloper-firstEnterprise-ready

Pros

  • ✓Genuinely open source under Apache 2.0, so the agent's prompting and tool-use logic can be read, forked and audited — several HN commenters said they learned agent design from the codebase
  • ✓No vendor lock-in on models: the same agent runs against frontier APIs, your own Bedrock or Vertex contract, or a local model through Ollama
  • ✓Enterprise deployment options run to VPC, on-premises and air-gapped, which is rare among coding agents and decisive for regulated buyers
  • ✓Strong practitioner reputation for long autonomous runs — reports of planning for ten to fifteen minutes then producing an accurate multi-file diff unattended

Cons

  • ✗Token-hungry by design: it sends extensive context with each request, and independent 2026 comparisons put active frontier-model sessions at roughly $5-15 per hour, which is unpredictable against a flat $20/month competitor
  • ✗No codebase indexing is a real limitation on very large repositories, and HN critics maintaining tens of thousands of reference files call semantic search a killer feature Cline is missing
  • ✗Multiple practitioners on Hacker News report Claude Code reaching a correct result in fewer prompts on complex tasks
  • ✗A compromised npm publish token let an attacker ship a malicious Cline CLI 2.3.0 for about eight hours on 17 February 2026, downloaded roughly 4,000 times; the root cause was prompt injection into the project's own AI-powered issue-triage workflow, since removed in favour of OIDC provenance publishing
  • ✗Accuracy is reported to degrade once context grows past a few tens of thousands of tokens, so very large tasks still need decomposition

Pricing

Open Source

$0

  • ✓Full VS Code extension and CLI
  • ✓Bring your own API keys
  • ✓MCP marketplace access
  • ✓Client-side architecture
  • ✓Pay only for model inference

Teams

From $20/user/mo

  • ✓Centralised billing and usage tracking
  • ✓Team management dashboard
  • ✓Budgets and access controls
  • ✓First seats free per Sacra's reporting

Enterprise

Contact for pricing

  • ✓SSO and OIDC
  • ✓Role-based access control and audit logs
  • ✓Global model and MCP policy enforcement
  • ✓VPC, on-premises and air-gapped deployment
  • ✓JetBrains extension
  • ✓SLA and dedicated support

The agent itself is free and Apache-2.0 licensed; the real cost is model inference, which you pay either directly through your own provider keys or through Cline's provider at cost with no markup — an explicit contrast with coding tools that embed margin in token usage. That makes it cheaper than a flat subscription for light users and more expensive for heavy ones: independent 2026 comparisons put active sessions with frontier models at roughly $5-15 per hour and typical monthly API spend between $25 and $200. Teams is reported at $20 per user per month, and Enterprise pricing for SSO, audit logging and self-hosted deployment is not published.

Security & Compliance

✗soc2
✗gdpr
✗hipaa
✗iso27001
✓sso
✓data residency

Connect

Sources

This page was written from 7 sources, 4 on domains other than cline.bot.

  1. 1.cline.bot — pricingvendor
  2. 2.cline.bot — enterprisevendor
  3. 3.cline.bot — cline raises 32m series a and seed funding building the openvendor
  4. 4.github.com — cline
  5. 5.sacra.com — cline
  6. 6.news.ycombinator.com — item
  7. 7.thehackernews.com — cline cli 230 supply chain attack
Newsletter

Stay Ahead of the Curve

Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.

Subscribe

Mentioned In

AI Governance

Pro-Human AI Declaration: A Roadmap for AI, If Anyone Will Listen

Enterprise AI analysis: Pro-Human AI Declaration. Strategic insights, ROI considerations, and implementation guidance for technical and business leaders eval...

March 13, 2026
Enterprise AI

Anthropic Wins 70% of Enterprise Deals—OpenAI's Losing Grip

70% of enterprise AI deals go to Anthropic, not OpenAI. For CFOs: the cost model shift driving vendor consolidation and what it means for 2026 budgets.

March 14, 2026
Enterprise AI

Salesforce's 26% Drop: The Death of Seat-Based Pricing

Salesforce's 26% Drop. For CFOs and finance leaders: cost implications, budget planning, and ROI benchmarks from enterprise AI deployments.

March 16, 2026
OpenAI

OpenAI Offers 17.5% Returns to Win Enterprise AI Battle Against Anthropic

OpenAI Offers 17.5% Returns to Win Enterprise AI Battle Against Anthropic. For enterprise decision-makers: strategic analysis, cost implications, and impleme...

March 23, 2026
OpenAI

OpenAI's 17 Acquisitions: Vendor Lock-In and $207B Funding Gap

OpenAI's 17 Acquisitions in 3 Years Signal Vendor Lock-In Risk as $207B Funding Gap Looms. For enterprise decision-makers: strategic analysis, cost implicati...

March 25, 2026
Apple

Apple Siri 2.0 at WWDC 2026 Targets Enterprise With Autonomous AI Agents and Post-Quantum Security

Apple Siri 2.0 at WWDC 2026 Targets Enterprise With Autonomous AI Agents and Post-Quantum Security. For enterprise decision-makers: strategic analysis, cost ...

March 26, 2026
OpenAI

OpenAI $25B Revenue: Why Late 2026 IPO Tests Markets

Analysis of OpenAI $25B Revenue. For enterprise leaders: strategic implications, cost considerations, and implementation guidance for AI decision-makers.

April 4, 2026

Latest Articles

View All →