Plandex
Plandex is an open-source, terminal-based AI coding agent designed to take on large, multi-step software development tasks across many files in real world codebases. Written in Go and released under the MIT license, Plandex builds and executes long-running "plans" — durable, branchable units of work that combine intelligent context management, project maps generated with tree-sitter for 30+ programming languages, an effective 2M-token context window, and a cumulative diff review sandbox that isolates AI edits until a developer explicitly applies them. The tool ships as a CLI with an interactive REPL (fuzzy auto-complete), supports automated debugging of terminal commands and browser applications, integrates with Git for branching and commit-message generation, and is provider-neutral — working with Anthropic Claude, OpenAI, Google Gemini, OpenRouter.ai, Azure OpenAI, AWS Bedrock, DeepSeek, Perplexity, Ollama, and any OpenAI-compatible custom provider. The Plandex Server exposes a REST management/orchestration API (over 60 endpoints across accounts, orgs, projects, plans, branches, context, conversation, diffs, settings, model packs, and streaming execution) that powers both the CLI/REPL and the hosted Plandex Cloud. Plandex Cloud is winding down as of 2025-10-03; Plandex is now distributed primarily as a Docker-based self-hosted / local-mode product that users run with their own model-provider API keys.
Solid contracts, transparent operations, and an easy start — typically complete on four or five facets with one clear soft spot.
API Evangelist profiles Plandex the way a machine reads it — 87 machine-readable artifacts across 15 APIs, pulled from the provider's own public surface and indexed so a developer, an analyst, or an AI agent can evaluate it against every other provider on the network.
Every provider in the network is reduced to the same set of machine-readable artifacts — OpenAPI contracts, event specifications, GraphQL schemas, runnable collections, pricing and rate-limit signals, security posture, OAuth scopes, and the agent surfaces (MCP servers and skills) that let software drive the API on its own. We profile them because the interface is the part of a company you can actually inspect: it is a truer signal of what a provider does than any marketing page. From those artifacts we compute the Kin Score — Plandex scores 67.1/100 (strong), with a separate agent-readiness read of 48/100 (agent ready). The full breakdown is below, followed by every artifact we hold — each card links through to its machine-readable definition on apis.io.
Kin Score
This is the API Evangelist rating — a single, repeatable read computed from the artifacts on this page. Green fill is points earned; the red track is points possible, so every bar shows earned-versus-possible at a glance.
Kin Score
How this is scored →scored 2026-07-27 · rubric v0.5
How we profile Plandex
Each block below is one kind of artifact we hold for Plandex. For each we say what it is and why it earns a place in the profile, then list every one we've indexed — capped at two rows, scroll within the panel for the rest.
APIs 15
Each API is captured as its own OpenAPI definition — every operation, parameter, and response. This is the single most useful machine-readable description of what an API does, and it's what lets us score, lint, mock, and generate against it without asking the provider for anything.
Individual APIs this provider publishes, each with its own machine-readable definition.
Plandex Accounts API
User account, email verification, and authentication operations.
Plandex Branches API
Per-plan branches for parallel exploration and diffing.
Plandex Context API
Files, directory trees, URLs, notes, and images loaded into a plan.
Plandex Conversation API
Prompts, responses, and rewind history within a plan branch.
Plandex Diffs API
Per-file pending changes, apply, and reject operations.
Plandex Execution API
Streaming connect, build, tell, and stop endpoints for plan execution.
Plandex FileMap API
Tree-sitter project map generation and cached map loading.
Plandex Health API
Operational health and version endpoints.
Plandex Invites API
Invite users into an org and manage pending invites.
Plandex Models API
Model packs, custom models, custom providers, default settings.
Plandex Orgs API
Organization and role membership management.
Plandex Plans API
Long-running, branchable units of AI coding work over loaded context.
Plandex Projects API
Project containers that group plans, mapped to a working directory.
Plandex Settings API
Per-plan model settings, default settings, and org/user config.
Plandex Users API
User listing and removal within an org.
Scroll within the panel for all 15 ·
Open Collections 1
Open, tool-agnostic collections carry the same runnable value as Postman without locking you to one client — the portable, forkable form of the same exercise.
Open, tool-agnostic API collections (OpenAPI-derived and Bruno).
Plandex Server API
OPEN COLLECTIONPricing Plans 1
Pricing is part of the interface. Machine-readable plans tell you what a tier costs and includes before you commit — one of the six things the Kin Score reads for commercial clarity.
Published pricing tiers and plan structures.
Rate Limits 1
Rate limits are the difference between a demo that works and a production integration that doesn't fall over. Publishing them is an operational-transparency signal — and a hard requirement for any agent that plans its own throughput.
Documented rate limits and quota policies.
Plandex Rate Limits
RATE LIMITSFinOps 1
Cost, billing, and metering signals let a buyer model the financial operations of an API before it's live. We profile them for the same reason we profile pricing: the money is part of the contract.
Cost, billing, and metering signals for API financial operations.
Plandex Finops
FINOPSFeatures 16
The notable capabilities this provider advertises, captured as structured features so they can be searched and compared instead of read one landing page at a time.
Notable capabilities this provider offers.
2M-Token Effective Context Window
Smart context management loads only what's needed per step, enabling reliable work in large projects and files.
Tree-Sitter Project Maps
Fast project map generation and syntax validation across 30+ programming languages. Indexes directories with 20M+ tokens.
Cumulative Diff Sandbox
AI-generated changes are isolated from project files until explicitly applied, with cumulative review and roll-back.
Configurable Autonomy
Five autonomy levels — None, Basic, Plus, Semi-Auto, Full-Auto — selectable per-plan or via CLI flags.
Automated Debugging
Repeatedly runs commands (builds, tests, lints, scripts) and auto-fixes failures with rollback. Also debugs browser apps via Chrome.
Model Packs
Curated combinations of models bound to internal roles (planner, coder, builder, summarizer, verifier, context loader); built-in packs for daily, reasoning, strong, cheap, oss, ...
Multi-Provider Support
First-class support for Anthropic, OpenAI, Google AI Studio, Google Vertex AI, Azure OpenAI, AWS Bedrock, DeepSeek, Perplexity, OpenRouter, Ollama, and any OpenAI-compatible cus...
Claude Pro/Max Subscription
Plandex can use a Claude Pro or Max subscription as the credential when calling Anthropic models.
Context Caching
Built-in context caching across OpenAI, Anthropic, and Google models reduces latency and cost on multi-step plans.
Plan Version Control
Every plan update is versioned, with branching for exploring multiple paths or comparing different models. Rewind to any prior SHA.
Git Integration
Commit message generation, optional auto-commits, and clean integration with project git history.
REPL with Fuzzy Auto-Complete
Interactive shell launched by `plandex` (or `pdx`) with fuzzy command and file completion; backslash equivalents (`\new`, `\tell`) for every command.
One-Line Install
Zero-dependency CLI install via `curl -sL https://plandex.ai/install.sh | bash`. Docker compose for the self-hosted server.
Background Tasks
`--bg` flag runs plans concurrently in the background under separate streams.
Smart Context + Auto-Load Context
Loads only the files needed for each implementation step (smart-context) and uses the project map to choose them (auto-load-context).
Streaming Plan Execution
tell, build, and connect endpoints stream model responses live; clients can reconnect to in-progress plans after disconnects.
Scroll within the panel for all 16 ·
Semantic Vocabularies 1
JSON-LD contexts give the data shared meaning across APIs. We profile them because semantics are what let a machine reconcile 'customer' here with 'customer' somewhere else.
JSON-LD contexts and semantic vocabularies used across these APIs.
Plandex Context
JSON-LDSpectral Rules 2
Governance rulesets we run against this provider's specs — the automated checks behind parts of the score. Profiling them makes the quality bar explicit and re-runnable, not a matter of opinion.
Plandex API Rules
SPECTRALPlandex API Rules
SPECTRALJSON Schema 6
Standalone JSON Schema definitions describe the data models behind the API. We profile them so the shapes are validatable on their own — useful long after a single request is forgotten.
Standalone JSON Schema definitions for this provider's data models.
Branch
JSON SCHEMAContextItem
JSON SCHEMAConvoMessage
JSON SCHEMAModelPack
JSON SCHEMAPlanConfig
JSON SCHEMAPlan
JSON SCHEMAJSON Structure 3
JSON Structure captures the data shapes in a form built for tooling — a complement to JSON Schema that keeps the model machine-legible.
JSON Structure definitions describing this provider's data shapes.
Plandex Server Context Item Structure
JSON STRUCTUREPlandex Server Model Pack Structure
JSON STRUCTUREPlandex Server Plan Structure
JSON STRUCTUREExamples 5
Real request and response payloads are what turn a spec from abstract into obvious — and they're one of the twelve things an agent needs to call an API correctly on the first try.
Example request and response payloads for these APIs.
Security Posture 1
Authentication, domain security, vulnerability disclosure, and trust-center signals — the evidence that a provider takes security seriously enough to document it. We profile it because you can't govern what you can't see.
Authentication, domain security, vulnerability disclosure, and trust-center signals.
Agentic Access 1
An x-agentic-access contract marks which operations are safe for an agent to run on its own and which need a human in the loop. It is the difference between an API an agent can use and one it can use safely.
Recommended x-agentic-access execution contracts for AI agents.
Use Cases 9
What developers actually build with this provider — captured so the catalogue answers 'what is this for', not just 'what does this expose'.
What developers build with this provider.
Large Refactors
Coordinate refactors that touch dozens of files across a large codebase without losing the thread between steps.
Polyglot Project Work
Work in repositories that span 30+ languages thanks to tree-sitter project maps.
Adding Features To Real Projects
Take a feature from idea through chat mode, into a detailed implementation plan, into reviewed and applied file changes.
Automated Debugging
Wrap a flaky build, test, or browser app in `plandex debug` and let Plandex iteratively repair it.
Codebase Onboarding And Q&A
Use chat mode and the project map to ask questions about an unfamiliar codebase.
Multi-Model Comparisons
Branch a plan to run different model packs against the same context and compare results.
Long-Running Background Coding
Kick off `--bg` plans, switch repos, and reconnect to streaming output later.
BYO Subscription Coding Agent
Use a Claude Pro/Max subscription as the credential for an autonomous coding agent.
Self-Hosted AI Coding For Sensitive Codebases
Run the Plandex server in Docker on-prem with locally-hosted Ollama models or a private custom provider.
Scroll within the panel for all 9 ·
Integrations 21
Pre-built integrations with other platforms tell you where this provider already fits in a stack.
Pre-built integrations with other platforms and tools.
Anthropic Claude
Direct Anthropic API integration; also supports Claude Pro/Max subscriptions.
OpenAI
Direct OpenAI API integration via OPENAI_API_KEY.
OpenRouter.ai
Default recommended provider for self-hosted Plandex; also used as a failover route across other configured providers.
Google AI Studio
Gemini models via GEMINI_API_KEY.
Google Vertex AI
Gemini and Anthropic models via Vertex AI credentials.
Microsoft Azure OpenAI
OpenAI models on Azure with deployment-name mapping.
AWS Bedrock
Anthropic models on Bedrock with AWS profile or env-var credentials.
DeepSeek
DeepSeek models via DEEPSEEK_API_KEY.
Perplexity
Perplexity models via PERPLEXITY_API_KEY.
Ollama
Local model hosting via Ollama; no API keys required.
Custom Providers
Any OpenAI-compatible API can be added via a JSON config (`plandex models custom`).
Relace
Instant-apply fallback model used when Plandex cannot apply edits deterministically (Plandex Cloud).
Git
Branching, commit-message generation, optional auto-commits.
Chrome
Automated debugging of browser applications when Chrome is installed.
Docker
Self-hosted server distributed as a Docker image with a published docker-compose stack.
Postgres
Backing database for the Plandex server.
LiteLLM Proxy
Embedded LiteLLM proxy used by the server to talk to model providers uniformly.
Stripe
Payment processing for Plandex Cloud subscriptions and credit purchases.
Loops
Email marketing for Plandex updates (Plandex Cloud).
Google Analytics
Basic usage analytics (Plandex Cloud).
Rollbar
Error tracking (Plandex Cloud).
Scroll within the panel for all 21 ·
Solutions 3
Packaged solutions the provider offers on top of the raw API surface.
Packaged solutions this provider offers.
Self-Hosted / Local Mode
Run Plandex locally with Docker and BYO model-provider keys. MIT-licensed and free.
Plandex Cloud — BYO API Key Mode (Historical)
Hosted Plandex with user-supplied provider keys, $30/month after trial. Winding down 2025-10-03.
Plandex Cloud — Integrated Models Mode (Historical)
Hosted Plandex with integrated credit billing, $45/month including $20/mo of non-expiring credits. Winding down 2025-10-03.
Resources
Every other property we hold for Plandex — documentation, portals, status pages, policies, and corporate surface — grouped by the job it does, following the integrator's arc from getting started to running in production.
Get Started 2
Portal, sign-up, and the first successful call
Documentation 1
Reference material describing how the API behaves
Agent Surfaces 1
MCP servers, agent skills, and machine-readable catalogs
Design & Contract 3
Pagination, idempotency, versioning, errors, and events
Build 5
SDKs, sample code, and the tooling you integrate with
Access & Security 2
Authentication, authorization, and security posture
Learn 1
Tutorials, courses, talks, and written guidance
Operate 7
Status, limits, changes, and where to get help
Scroll within the panel for all 7 ·
Commercial 6
Pricing, plans, and the legal terms of use
Company 3
The organization behind the API
Other 4
Properties that don't map to a standard resource type
← All providers · Data indexed from github.com/api-evangelist/plandex · machine-readable index on apis.io