# Verdent (slim)

> Verdent is a coding agent from Verdent AI, Inc., with a desktop app for macOS and Windows, VS Code and JetBrains extensions and a browser edition. It plans tasks, runs parallel workers in git worktrees and uses MCP servers.

- Full: https://www.anchorterminal.com/tools/verdent.md (~8,100 tokens) · this version ~1,430 tokens · JSON https://www.anchorterminal.com/tools/verdent.json · canonical https://www.anchorterminal.com/tools/verdent
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-08 · pricing https://www.anchorterminal.com/tools/verdent#pricing · reviews https://www.anchorterminal.com/tools/verdent#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**E · 39.7/100 · rank #1339 of 1513 · #53 in Agent harnesses · not agent-ready · confidence medium**


Assessment: A desktop and editor coding agent that splits a goal into parallel workers in git worktrees, with a read-only Plan Mode, MCP servers and the option to run workers on Claude Code or Codex. It has no CLI, API or sandbox, sends context through Verdent's servers, and has had no free tier since 29 October 2026.

## Facts

- Kind: Agent harness · vendor: Verdent AI, Inc. · category: Agent harnesses · legal entity: Verdent AI, Inc. · provenance 62/100
- Auth: OAuth or key · pricing: Paid · x402: no · licence: Proprietary. The terms of use from Verdent AI, Inc. forbid reverse engineering or copying any component, and no source is published
- Benchmark probes: not run yet, so Performance is pending
- Surface listed: Verdent Desktop (Manager), with the VS Code extension and browser edition named, read from verdent.ai (docs, changelog, pricing, security, terms, privacy and about pages) on 6 November 2026. Nothing was installed, no account was made and no task was run
- Interfaces: Desktop app for macOS 11 or later and 64-bit Windows 10 and 11 (no Linux), Verdent for VS Code, a JetBrains plugin named in the footer and terms, and Verdent Cloud in the browser, which publishes web apps to *.verdent.app
- Modes: Agent Mode (the default) reads, writes and runs commands, with prompts for destructive operations, edits outside the expected scope and some tools. Plan Mode is read-only and writes a plan.md for review
- Parallel work: Manager splits a goal into subtasks and dispatches workers in parallel, each in its own git worktree, with a review column on a Kanban board
- Headless: No CLI, print mode, API or SDK found. Work is started from the app or from a paired Telegram, Slack or Feishu bot
- Sandbox: No operating-system sandbox or network restriction documented. Workspace isolation is by git worktree, and the docs advise relying on git to roll back
- Data path: Requests carry conversation history and tool results to Verdent's AWS servers in the US, then to Azure, Bedrock or Vertex. Web search goes to Parallel.ai and page fetches to Jina. Event logging goes to log.verdent.ai, with no opt-out found
- Training: The privacy policy says input and output are not used for training without explicit consent. The security page refers to a Privacy Mode that, when off, allows training
- Releases: 1.0.0 on 23 September 2025, 2.0.0 on 17 April 2026 and 2.18.0 on 29 October 2026, per the dated changelog
- Prices: Lite plan $5 per month (plan); Starter plan $19 per month (plan); Pro plan $59 per month (plan); Max plan $179 per month (plan); Credit $0.059 per credit

- Scores: Reliability 43, Performance pending, Schema & documentation 55, Agent ergonomics 35, Security & auth 33, Payments & pricing 20, Task success pending, Maintenance & community 61, Transparency & trust 55 · negative events -2 · total over the 7 assessed categories
- Why: Reliability, Local-package reading, as for the other closed-source harnesses. · Schema & documentation, Framework reading for a harness. · Agent ergonomics, Harness reading. · Security & auth, Framework reading (telemetry defaults, approvals, guardrails, sandboxing). · Payments & pricing, Harness reading of the published rubric, scored on the Verdent subscription the agent needs. · Maintenance & community, Closed service. · Transparency & trust, Proprietary, under terms from Verdent AI, Inc.
- Sources: 15, open questions: 4, both in the full twin
- Capabilities: agent.harness, agent.multi-agent, agent.mcp-client
- JSON: https://www.anchorterminal.com/api/v1/tools/verdent.json
- Its vendor verifies the listing with `verify_listing` or `POST https://www.anchorterminal.com/api/v1/verify` (the full twin says how); no effect on the grade.
- Support: Discord https://discord.gg/NGjXEZcbJq; support@verdent.ai; https://www.verdent.ai/docs/verdent-manager/reference/support-resources; more in the full twin

## Before you call it

1. Don't plan to drive Verdent headless. Hand the task to the person using the desktop app, or use a harness with a print mode
2. Start large changes in Plan Mode and review plan.md before switching to Agent Mode
3. Add MCP servers in the project's .mcp.json or ~/.verdent/mcp.json in the mcpServers shape
4. Keep work under git, since rollback relies on git and there is no sandbox
5. Budget in credits at about $0.059 each, because there is no free allowance since 29 October 2026

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| goose | BB | 71.3 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/goose.min.md |
| AXLR | BB | 70.6 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/axlr.min.md |
| OpenHands | BB | 70.3 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/openhands.min.md |
| OpenCode | B | 68.8 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/opencode.min.md |
| Qwen Code | B | 68.7 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/qwen-code.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)