"
```
## Related Resources
* **[Advanced Capabilities](/work-with-devin/advanced-capabilities)** — Overview of Devin's advanced features
* **[Devin's MCP Marketplace](/work-with-devin/mcp)**
* **[Connecting remote MCP servers to Claude](https://support.anthropic.com/en/articles/11175166-about-custom-integrations-using-remote-mcp)**
* **[OpenAI's docs for using MCP servers](https://platform.openai.com/docs/guides/tools-remote-mcp)**
* **[DeepWiki MCP](/work-with-devin/deepwiki-mcp)** — For public repositories only
* **[DeepWiki](/work-with-devin/deepwiki)**
* **[Ask Devin](/work-with-devin/ask-devin)**
## 6. GET https://docs.devin.ai/admin/billing/self-serve.md
HTTP/2 200
age: 54
cache-control: public, max-age=0, must-revalidate
cf-cache-status: EXPIRED
cf-ray: a4102026a9dd1f5e-IAD
content-disposition: inline
content-security-policy: default-src 'none'
content-type: text/markdown; charset=utf-8
date: Sat, 26 Sep 2026 06:13:56 GMT
last-modified: Sat, 26 Sep 2026 06:13:56 GMT
link: ; rel="llms-txt", ; rel="llms-full-txt", ; rel="api-catalog", ; rel="mcp-server-card", ; rel="agent-card", ; rel="agent-skills"
server: Vercel
strict-transport-security: max-age=63072000
vary: rsc, accept-encoding
x-content-type-options: nosniff
x-frame-options: DENY
x-llms-txt: /llms.txt
x-matched-path: /_mintlify/_markdown/_sites/[subdomain]/[[...slug]]
x-robots-tag: noindex
x-vercel-cache: HIT
x-vercel-id: iad1:iad1:iad1::iad1::bf28q-1790403236881-5ab7ca7b6d00
[body: 8797 bytes as received, sha256 f81d106df8e18e64102de82f804c5fc4fc3ab0d2683c1dc9077f9e32a168bed8]
> ## Documentation Index
> Fetch the complete documentation index at: https://docs.devin.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Self-serve plans
> Compare Devin's Free, Pro, Max, and Teams self-serve plans, including seat types, the Teams minimum, usage quota, and on-demand credits.
Devin has four self-serve plans you can sign up for directly at [app.devin.ai](https://app.devin.ai). For authoritative pricing, see the [Devin pricing page](https://devin.ai/pricing). This page explains how the plans relate to each other and how billing mechanics work.
## Plan overview
| Plan | For | Price | Members |
| --------- | --------------------------------- | ------------------ | --------- |
| **Free** | Individuals trying Devin | Free | 1 |
| **Pro** | Individual users | \$20/month | 1 |
| **Max** | Power users who need more quota | \$200/month | 1 |
| **Teams** | Teams working with Devin together | \$80/month minimum | Up to 200 |
The **Pro** and **Max** plans are individual plans. They cannot be shared across multiple users. If you want multiple people to use Devin under a single subscription, you need the **Teams** plan.
## Free
The Free plan lets you try Devin with limited usage. It includes:
* Limited Devin usage
* Access to [Devin Review](/work-with-devin/devin-review)
* Access to [DeepWiki](/work-with-devin/deepwiki)
Free users can upgrade to any paid plan at any time from [Settings > Plans](https://app.devin.ai/settings/plans).
## Pro
Pro is Devin's entry-level individual plan. It's designed for a single developer who uses Devin regularly. Pro includes:
* A daily and weekly usage quota that covers Devin sessions, [Devin CLI](/cli), and [Devin Desktop](https://windsurf.com)
* Pay-as-you-go [on-demand credits](#on-demand-credits) for usage past your quota
* Slack, Linear, and [MCP](/work-with-devin/mcp) integrations
Pro is a single-user plan. Pro subscribers cannot invite additional members to their organization. To work with teammates on a shared subscription, use the [Teams](#teams) plan.
## Max
Max is for individual users who consistently exceed the Pro quota. It includes everything in Pro, plus a significantly larger weekly usage quota (with no daily cap), also shared between Devin sessions, [Devin CLI](/cli), and Devin Desktop.
Like Pro, Max is a single-user plan and does not support multiple members.
## Teams
The Teams plan is Devin's self-serve plan for teams of up to 200 users. Key properties:
* **Up to 200 members**: invite teammates until your organization reaches 200 users. Larger teams should [contact sales](https://cognition.com/contact) about Enterprise.
* **\$80/month minimum**: every Teams account pays at least \$80/month.
* Each member gets either a **full seat** or a **flex seat**.
* **On-demand credits** are shared across the whole team.
### Full seats vs. flex seats
Every member of a Teams account holds one of two seat types:
**\$40/month per seat**, billed as a fixed recurring line item.
Best for members who use Devin regularly. Each full seat includes:
* A daily and weekly usage quota equivalent to the Pro plan
* Access to Devin Desktop
**Free**. There is no separate limit on flex seats, but every member (full or flex) counts toward the 200-user cap.
Best for occasional users. Flex seats:
* Draw entirely from the team's shared pool of on-demand credits
* Do **not** include Devin Desktop access
* Have no fixed monthly charge per seat
Admins choose the seat type when [inviting a member](/product-guides/invite-team), and can convert a member between seat types later from **Settings > Members**.
### The \$80 Teams minimum
Every Teams subscription costs **at least \$80/month**. You can hit this minimum in any combination of full seats (\$40 each) and on-demand credits:
| Full seats | On-demand credits included | Monthly total |
| ---------: | -------------------------: | ------------: |
| 0 | \$80 | \$80 |
| 1 | \$40 | \$80 |
| 2 | \$0 | \$80 |
| 3 | \$0 | \$120 |
| *N* ≥ 2 | \$0 | *N* × \$40 |
When you have fewer than two full seats, the remainder of the \$80 minimum is automatically charged as prepaid on-demand credits that the whole team can draw from. Once you have two or more full seats, you've cleared the minimum and no additional on-demand credits are included, but you can still [top up on-demand credits](#on-demand-credits) at any time.
Give a full seat to anyone who uses Devin regularly. Full seats include their own Pro-equivalent quota and Devin Desktop access at a predictable fixed cost, making them the best fit for power users. Reserve flex seats for occasional or trial users who only need ad-hoc access through shared on-demand credits.
## How quotas work
Each paid plan and full seat includes a usage allowance that refreshes automatically on a calendar basis:
* **Pro** and **Teams full seats** have a **daily and weekly** allowance. The daily allowance is more than 1/7 of the weekly, so you can keep working through weekends without giving up overall capacity for the week.
* **Max** has a **weekly** allowance only, with no daily cap.
When you've used up your allowance, [on-demand credits](#on-demand-credits) keep you working without interruption.
## On-demand credits
On-demand credits are prepaid usage credit that fund any work past your plan's included quota:
* **Roll over** month-to-month. Purchased credits never expire.
* Can be topped up at any time from [Settings > Plans](https://app.devin.ai/settings/plans), and optionally refilled automatically via auto-reload.
* Admins can set auto-reload thresholds and default session spending limits from [Settings > Usage & limits](https://app.devin.ai/settings/usage).
* On the **Teams** plan, credits are **shared across all members**, with no per-member balance. Any teammate can draw from the shared pool.
* On the **Teams** plan, credits fund all usage on **flex seats** and any **full seat** usage past its included quota, and cover any portion of the [\$80/month minimum](#the-80-teams-minimum) not already covered by full seats.
## Devin Review and Automations Pricing
[Automations](/product-guides/automations) and [Devin Review](/work-with-devin/devin-review) are available on self-serve plans.
* **Teams use shared on-demand credits.** On the Teams plan, Automations and Devin Review draw directly from the team's shared [on-demand credit](#on-demand-credits) pool and do not consume full-seat quota.
* **What happens when you run out of credits.** If you run out of credits, Automations stop running and Devin Review switches to its smart diff viewer. Top up [on-demand credits](#on-demand-credits) to start Automations again and re-enable AI-powered review.
* **Public PRs are free.** Anyone can review a public GitHub PR at [devinreview.com](https://devinreview.com) — or by replacing `github.com` with `devinreview.com` in any PR URL — without a Devin account, and no on-demand credits are consumed.
Admins can keep usage predictable by tuning how often auto-review runs. Configure the trigger mode (every commit, only when a PR is first opened, or manual only) per repository or per user from [Settings > Review](https://app.devin.ai/settings/review). See [Trigger Modes](/work-with-devin/devin-review#trigger-modes) in the Devin Review docs for details.
## Migrating from legacy ACU-based plans
If you were previously on a legacy ACU-based plan, here's what you need to know:
* On-demand credits are the same dollar value as the ACUs you're used to.
* **Legacy Core plan users** have been migrated to the Free plan and can continue using any remaining on-demand credits. To purchase additional credits, upgrade to the [Teams](#teams) plan.
## Managing your plan
Admins can view and change the account's plan from [Settings > Plans](https://app.devin.ai/settings/plans). From there you can:
* Upgrade or downgrade between Free, Pro, Max, and Teams
* Add or cancel Teams full seats
* Purchase on-demand credits or configure auto-reload to replenish them automatically
* Download past invoices
For tips on keeping consumption under control across all plans, see [Usage](/admin/billing/usage).
## 7. GET https://docs.devin.ai/admin/billing/usage.md
HTTP/2 200
age: 0
cache-control: public, max-age=0, must-revalidate
cf-cache-status: EXPIRED
cf-ray: a410202b3fcae5c7-IAD
content-disposition: inline
content-security-policy: default-src 'none'
content-type: text/markdown; charset=utf-8
date: Sat, 26 Sep 2026 06:13:57 GMT
last-modified: Sat, 26 Sep 2026 06:13:58 GMT
link: ; rel="llms-txt", ; rel="llms-full-txt", ; rel="api-catalog", ; rel="mcp-server-card", ; rel="agent-card", ; rel="agent-skills"
server: Vercel
strict-transport-security: max-age=63072000
vary: rsc, accept-encoding
x-content-type-options: nosniff
x-frame-options: DENY
x-llms-txt: /llms.txt
x-matched-path: /_mintlify/_markdown/_sites/[subdomain]/[[...slug]]
x-robots-tag: noindex
x-vercel-cache: REVALIDATED
x-vercel-id: iad1:iad1:iad1::iad1::vhc7v-1790403237599-62eda037273b
[body: 4561 bytes as received, sha256 9503ab89e6c3f7456b40b757a2ca451460948df2efffa2465a62748aa441d4aa]
> ## Documentation Index
> Fetch the complete documentation index at: https://docs.devin.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Usage
> How Devin meters work, what counts toward consumption, and how to keep usage under control
This page explains how Devin's work is metered. The mechanics are the same regardless of pricing model. The only difference is the unit:
* **Enterprise** customers consume **Agent Compute Units (ACUs)** against the volume in their order form.
* **Self-serve** customers consume their plan's included quota first, then draw from prepaid **on-demand credits**.
Throughout this page, "usage" refers to whichever unit applies to your account.
## What counts toward usage
Usage accrues based on the work Devin actually performs in a session, including:
* Number and complexity of actions Devin takes (planning, context gathering, task execution, browser actions, code execution, and so on)
* Virtual machine time and networking bandwidth (typically a small fraction of total usage)
### Windows sessions
Windows sessions consume approximately **9% more** usage than equivalent Linux (Ubuntu) sessions.
### macOS sessions
[macOS sessions](/onboard-devin/environment/macos-support) currently consume the **same** usage as equivalent Linux (Ubuntu) sessions — there is no macOS surcharge.
This is **promotional launch pricing** and is subject to change.
Aside from the few units required to keep the Devin VM running, Devin will not consume usage when:
* Waiting for your response
* Waiting for a test suite to run
* Setting up and cloning repositories
## Sleep and idle behavior
When a session is idle, Devin goes to sleep. While sleeping, Devin does not consume usage. You can wake the session up at any time by sending another message. Devin sleeps automatically after 30 minutes of inactivity by default. Enterprise customers can ask their Cognition account team to adjust this timeout (between 5 and 120 minutes).
## Managing usage effectively
A number of variables affect how much Devin consumes:
* Task complexity
* Prompt quality (or specificity)
* Size of context or codebase
* Number of files being touched or modified
* Session runtime
* Length of conversation
* Frequency of back-and-forth messaging
A few tips to keep usage under control:
* Delegate clearly scoped tasks with a well-defined end goal
* Keep prompts and sessions short
* Avoid asking Devin to do a lot of different tasks in the same session
* Split big projects into sub-tasks across sessions; there are no concurrent session limits, so take advantage of it
These tips also tend to improve the quality of Devin's work, so it's a win-win. [Devin Coach](/enterprise/features/devin-coach) reinforces them automatically by flagging inefficient prompts in the input box before they're sent.
Enterprise admins can additionally cap each member's monthly consumption with [usage policies](/enterprise/features/usage-policies).
## Frequently asked questions
No, Devin does not consume any usage while sleeping.
Devin sleeps automatically after 30 minutes of inactivity by default, so awake-but-idle time generally adds up to very little.
Any user can see per-session usage from [Session Insights](/product-guides/session-insights), regardless of pricing model.
* **Self-serve**: Current month's usage, quota remaining, and on-demand credit balance live at [Settings > Usage & limits](https://app.devin.ai/settings/usage).
* **Enterprise**: Enterprise and per-Organization ACU consumption are available from the [Consumption](https://app.devin.ai/settings/consumption) and [Consumption Analytics](https://app.devin.ai/settings/analytics?tab=consumption) pages in Enterprise and Organization Settings respectively.
Yes. If your enterprise enables [Personal Analytics](/enterprise/security-access/personal-analytics), users with the **View Personal Analytics** permission can see their own ACU consumption across every organization from the **My analytics** page in their settings.
## 8. GET https://devin.ai/pricing (the server answered text/markdown to this user agent; body as received)
HTTP/2 200
accept-ranges: bytes
access-control-allow-origin: *
age: 28013
cache-control: public, max-age=0, must-revalidate
content-disposition: inline; filename="pricing.md"
content-type: text/markdown; charset=utf-8
date: Sat, 26 Sep 2026 06:13:18 GMT
etag: "0fdbdf32f443f979ccb75baad637822b"
last-modified: Fri, 25 Sep 2026 22:26:24 GMT
server: Vercel
strict-transport-security: max-age=63072000
vary: Accept, User-Agent, Signature-Agent
x-content-type-options: nosniff
x-frame-options: DENY
x-matched-path: /_md/pricing.md
x-vercel-cache: HIT
x-vercel-id: iad1::p6nqn-1790403198173-86f01a5b2d2b
content-length: 3961
[body: 3961 bytes as received, sha256 b7393ba902fdec8f9a067acc0c5306871af6fe86638106654a437742d1ccf0d0]
# Devin Plans and Pricing
Choose the perfect plan for your journey
[SWE-2, our latest model, is now available →](https://cognition.com/blog/swe-2)
INDIVIDUAL PLANS
Free
$0
[Download](https://devin.ai/download)
Light quota to code with agents
Limited model availability
Unlimited inline edits
Unlimited Tab completions
ProPOPULAR
$20per month
[Select plan](https://app.devin.ai/auth/signup)
Everything in Free, plus:
Increased quotas, including access to frontier models from OpenAI, Claude, Gemini, and SpaceXAI, plus leading open source models
Full model availability
Free use of SWE-2 Free in Devin Desktop and CLI through October 10, 2026
Access to cloud agents (Devin Cloud)
Purchase extra usage at API pricing
MaxNEW
$200per month
[Select plan](https://app.devin.ai/signup)
Everything in Pro, plus:
Significantly higher quotas
TEAM PLANS
Teams
$80/month for team plan + $40/mo per full dev seat
[Select plan](https://app.devin.ai/signup)
Everything in Pro, plus:
Up to 200 users
Share and collaborate
Centralized billing
Admin dashboard with analytics
Priority support
Enterprise
Let's talk
[Get a Demo](https://devin.ai/demo)
Built to scale across your organization:
Highest priority support
Dedicated account management
SAML/OIDC SSO
Centralized enterprise admin controls
Dedicated deployment option
First class support for every major model provider
## Compare Plans
Free$0/month
Pro$20/month
Max$200/month
Team$80/month
EnterpriseLet's talk
Members
1
1
1
Up to 200 users$40/month per full userEach full user includes their own generous quota and access to Devin Desktop
Unlimited
Devin Desktop
Tab
Unlimited
Unlimited
Unlimited
Unlimited
Unlimited
Premium models
Fast context
Free SWE-2Free in Devin Desktop and CLI through October 10, 2026
Slow
Devin Cloud
DeepWiki
Devin API
Ask Devin
Advanced capabilities
Sharing and collaboration
Concurrent sessions limit
Up to 10
Up to 10
Unlimited
Unlimited
Integrations
Slack and Teams
Linear and Jira
GitHub, GitLab, and Bitbucket
Custom git provider support
Account support
Dedicated Slack Connect channel for support
Optional onboarding call with the Cognition team
Dedicated account and engineering support
Access to early feature releases and research previews
Custom
Dedicated account team
Custom terms
Enterprise accountsCentralized management of multiple Devin organizations
Security
Deploy in your virtual private cloud (VPC)
SAML/OIDC SSO
Centralized enterprise admin controls
Teamspace isolation
## Frequently Asked Questions
How does usage work?
Each paid plan comes with a usage allowance that refreshes automatically on a daily and weekly basis. For the majority of users, this quota will be enough to fully cover all agent usage. Cost per message varies based on the model used, the task size and complexity, and the reasoning required.
What happens when I reach my usage limit?
Most users will never hit their limits. If you do go beyond your included usage on a paid plan, you can purchase extra usage which is consumed at API pricing.
How can I make my usage limits last longer?
**Control the size of your prompts.** Be precise with your instructions, but remove unnecessary context.\
\
**Switch to smaller models for routine tasks.** Using mini models like Haiku, GPT 5.2 Mini, and open source models like Kimi K2.5 will extend your usage.
What is the Max plan?
The Max plan is for power users who need significantly higher usage quotas. It includes everything in Pro, plus significantly higher quotas.
What is included in the Free plan?
The Free plan includes a light quota to code with agents, limited model availability, unlimited inline edits, and unlimited Tab completions.
## 9. GET https://deepwiki.com/.well-known/public-agents.json (still the application's HTML; stripped)
HTTP/2 200
age: 0
cache-control: public, max-age=0, must-revalidate
content-security-policy: frame-ancestors 'self' vscode-webview://*
content-type: text/html; charset=utf-8
date: Sat, 26 Sep 2026 06:12:39 GMT
etag: "odmc364yy2of4"
server: Vercel
strict-transport-security: max-age=63072000
vary: rsc, next-router-state-tree, next-router-prefetch, next-router-segment-prefetch
x-matched-path: /[org]/[repo]/[[...wikiRoutes]]
x-nextjs-prerender: 1
x-nextjs-stale-time: 300
x-powered-by: Next.js
x-vercel-cache: MISS
x-vercel-id: iad1::iad1::kjktw-1790403159564-2d5cf5721a33
content-length: 31648
[body: 31648 bytes as received, sha256 cb295df8bdf1501fb92902122c3e37cce7b836d0763c346d01ecb145b7f5ae1d]
.well-known/public-agents.json | DeepWiki
[aria-label=Notifications alt+T]
Index your code with Devin
DeepWiki
DeepWiki
[title=Open repository] .well-known/public-agents.json
Index your code with
Devin [aria-label=Share] Share
Loading...
## 10. GET https://cognition.ai/ (301 to cognition.com, as on 2026-09-08)
HTTP/2 301
cache-control: public, max-age=0, must-revalidate
content-type: text/plain
date: Sat, 26 Sep 2026 06:12:37 GMT
location: https://cognition.com/
server: Vercel
strict-transport-security: max-age=63072000
x-vercel-id: iad1::82mjb-1790403157115-b7d2b95cf599
[body: 15 bytes as received, sha256 7c040f8633b8823d72ed63da9b3b2dfe9846e912c66a64c9433ec9c4815d76c2]
Redirecting...
## 11. GET https://cognition.ai/terms followed: cognition.com/terms 301, cognition.ai/terms 301, lands on cognition.com/legal/website-terms-of-use 200 (Website Terms of Use, stripped)
HTTP/2 200
accept-ranges: bytes
access-control-allow-origin: *
age: 54438
cache-control: public, max-age=0, must-revalidate
content-disposition: inline
content-type: text/html; charset=utf-8
date: Sat, 26 Sep 2026 06:13:14 GMT
etag: "a51d82030ca2208a12e025bd6b7a1e2d"
server: Vercel
strict-transport-security: max-age=63072000
vary: rsc, next-router-state-tree, next-router-prefetch, next-router-segment-prefetch
x-content-type-options: nosniff
x-frame-options: DENY
x-matched-path: /legal/website-terms-of-use
x-nextjs-prerender: 1
x-nextjs-stale-time: 300
x-vercel-cache: HIT
x-vercel-id: iad1::s7d4d-1790403194058-e82073293a2b
content-length: 120489
[body: 120489 bytes as received, sha256 2e432c4c249c81438c8f2ab27636882d140fbb1841240cd3545a4c7f9ac78253]
Website Terms of Use | Cognition
[aria-label=Cognition home] [aria-label=Cognition]
Menu
[aria-label=Cognition home] [aria-label=Cognition] Close [aria-label=Primary]
Home
Careers
Research
Blog
Get a Demo
Devin
[aria-label=Primary] [aria-label=Cognition home] [aria-label=Cognition]
Home
Careers
Research
Blog
Get a Demo
Devin
Website Terms of Use
Last updated: May 13, 2024
The following terms and conditions (“Terms”) govern your access to and use of Cognition AI's website (“Cognition”, “we”, “us”, “our”), https://cognition.com/ (the “Site”). These Terms constitute a legally binding agreement made between you, whether personally or on behalf of an entity (“you”) and Cognition, concerning your access to and use of the website as well as any other media form, media channel, mobile website or mobile application related, linked, or otherwise connected thereto (collectively, the “Site”). Your use of Cognition's AI services or platform is governed by the separate Cognition Platform Terms of Use.
1. Acceptance of the Terms
You agree that by accessing the Site, you have read, understood, and agree to be bound by all of these Website Terms. IF YOU DO NOT AGREE WITH ALL OF THESE WEBSITE TERMS, THEN YOU ARE EXPRESSLY PROHIBITED FROM USING THE SITE AND YOU MUST DISCONTINUE USE IMMEDIATELY. You may find a copy of our Privacy Policy.
2. Changes to the Terms
We may modify, amend, supplement, or otherwise change the Terms from time to time at our sole discretion. All changes will be effective as of the “Last Updated” date stated at the top of this page. Each time you access the website, you consent to be bound by the most recently modified version of the Terms.
3. Your use of the Site
3.1 Intellectual property rights
Unless otherwise indicated, the Site is our proprietary property and all functionality, website designs, audio, video, text, photographs, and graphics on the Site (collectively, the “Content”) and the trademarks, service marks, and logos contained therein (the “Marks”) are owned or controlled by us, or licensed to us, or are used under applicable trademark nominative or fair use doctrines, and are protected by copyright and trademark laws and various other intellectual property rights and unfair competition laws of the United States, international copyright laws, and international conventions. The Content and the Marks are provided on the Site “AS IS” for your information and personal use only. No part of the Site and no Content or Marks may be copied, reproduced, aggregated, republished, uploaded, posted, publicly displayed, encoded, translated, transmitted, distributed, sold, licensed, or otherwise exploited for any commercial purpose whatsoever, without our express prior written permission. We reserve all rights not expressly granted to you in and to the Site, the Content and the Marks.
3.2 Restrictions
You agree not to use the Site in any way that violates any applicable federal, state, local, national, or international laws or regulations. We reserve the right to investigate complaints or reported violations of these Terms and to take any action as deemed appropriate by us, including referral to law enforcement or regulators for any illegal or unauthorized use of the Site, or terminating or blocking your access to the Site for any reason, with or without notice to you.
3.3 Availability
We reserve the right to modify and change the Site in our sole discretion without notice. No guarantees are made with respect to the Site's quality, stability, uptime, or reliability, and we will not be liable if for any reason all or any part of the Site is unavailable at any time for any period.
4. Warranty disclaimers
YOUR USE OF THE SITE IS AT YOUR OWN RISK. THE SITE, ITS CONTENT, AND ITEMS OBTAINED THROUGH THE SITE ARE PROVIDED ON AN “AS-IS”, AND “AS AVAILABLE” BASIS, WITHOUT ANY WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED. NEITHER COGNITION NOR ANY PERSON ASSOCIATED WITH COGNITION MAKES ANY WARRANTY OR REPRESENTATIONS WITH RESPECT TO THE COMPLETENESS, SECURITY, RELIABILITY, QUALITY, ACCURACY, OR AVAILABILITY OF THE SITE. WITHOUT LIMITING THE FOREGOING, NEITHER COGNITION NOR ANYONE ASSOCIATED WITH COGNITION REPRESENTS OR WARRANTS THAT THE SITE, ITS CONTENT, OR ANY ITEMS OBTAINED THROUGH THE SITE WILL BE ACCURATE, RELIABLE, ERROR-FREE, OR UNINTERRUPTED, THAT DEFECTS WILL BE CORRECTED, THAT OUR SITE OR THE SERVER THAT MAKES IT AVAILABLE ARE FREE OF VIRUSES OR OTHER HARMFUL COMPONENTS OR THAT THE SITE OR ANY ITEMS OBTAINED THROUGH THE SITE WILL OTHERWISE MEET YOUR NEEDS OR EXPECTATIONS. COGNITION HEREBY DISCLAIMS ALL WARRANTIES OF ANY KIND, WHETHER EXPRESS OR IMPLIED, STATUTORY OR OTHERWISE, INCLUDING BUT NOT LIMITED TO ANY WARRANTIES OF MERCHANTABILITY, NON-INFRINGEMENT, AND FITNESS FOR PARTICULAR PURPOSE.
5. Limitation of liability
YOU AGREE, TO THE MAXIMUM EXTENT PERMITTED UNDER APPLICABLE LAW, NOT TO HOLD COGNITION OR OUR SHAREHOLDERS, OFFICERS, DIRECTORS, EMPLOYEES, AGENTS, REPRESENTATIVES, SUPPLIERS, PARTNERS, AFFILIATES, OR LICENSEES RESPONSIBLE OR LIABLE FOR ANY DAMAGES OR REMEDIES OF ANY KIND ARISING FROM YOUR USE OF THE SITE, OR FROM A DELAY OR INABILITY TO ACCESS OR USE THE SITE (INCLUDING, BUT NOT LIMITED TO, YOUR RELIANCE UPON THIRD PARTY INFORMATION, CONTENT, VIRUSES, ERRORS, LINKED SITES, AND SERVICES OBTAINED THROUGH THE SITE) INCLUDING BUT NOT LIMITED TO ANY INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, PUNITIVE, LOST PROFITS, LOST GOODWILL, LOST DATA, WASTE OF EXPENDITURES, COST OF SUBSTITUTE, AND CONSEQUENTIAL DAMAGES, EVEN IF WE WERE EXPRESSLY ADVISED ABOUT THE POSSIBILITY OF SUCH DAMAGES. THIS LIMITATION APPLIES WHETHER THE DAMAGES ARE BASED IN CONTRACT, TORT, NEGLIGENCE, STRICT LIABILITY, OR OTHERWISE.
6. Indemnification
You agree to defend, indemnify, reimburse, and hold harmless Cognition as well as our shareholders, directors, officers, employees, agents, advisors, subsidiaries, affiliates, assignors, and licensors from and against any and all claims, causes of action, actions or demands, liabilities, damages, losses, fines, penalties, settlements, and costs and expenses of any kind (including reasonable attorney's fees), arising from or related to (i) your unauthorized use of the Site, (ii) your breach of any provision in these Terms, (iii) any of your representations or warranties made herein, or (iv) your violation of any law or rights of any third party. We will have the option to assume the exclusive defense and control of any action to which Cognition is named a party and you agree to cooperate with us in asserting any available defenses we find appropriate.
7. Dispute resolution
7.1 Arbitration
YOU AGREE THAT ALL MATTERS AND DISPUTES ARISING OUT OF OR RELATING TO THESE TERMS (“DISPUTES”) WILL BE RESOLVED THROUGH FINAL AND BINDING ARBITRATION, AND WILL BE GOVERNED BY THE LAWS OF THE STATE OF CALIFORNIA, USA, EXCLUDING CALIFORNIA'S CONFLICTS OF LAWS RULES.
The complaining party shall first attempt in good faith to resolve the Dispute by delivering written notice to the other party describing the facts and circumstances (with relevant documentation) of the Dispute, and allowing the receiving party thirty (30) days since the date of notice delivery to respond. If the Dispute is not resolved within thirty (30) days after the notice is delivered, the Dispute shall be resolved by arbitration by the American Arbitration Association's International Centre for Dispute Resolution in accordance with its Expedited Commercial Rules in force as of the date of these Terms (“Rules”). The parties will mutually select one arbitrator, and the arbitration shall be conducted in Santa Clara County, California, USA. The arbitration award will be final and binding on the parties and its execution may be presented in any competent court, including any court with jurisdiction over either party or any of its property. Each party will bear its own lawyer's and expert's fees and expenses, regardless of the arbitrator's final decision regarding the Dispute. Either party may apply to any competent court for injunctive or other equitable relief necessary to protect its rights pending resolution of the arbitration.
7.2 Limitations
Each claim involving a dispute must be brought to arbitration within twelve (12) months after the date on which such claim arose or it will be irrevocably waived. You hereby irrevocably waive your right to a jury trial as well as your right to claim that Santa Clara County, California, USA is an inconvenient or improper forum or venue to hear and resolve all claims or disputes concerning this agreement.
8. General provisions
8.1 Waiver and severability
No waiver by us of any term or condition set forth in these Terms shall be deemed a further or continuing waiver of such term or condition or a waiver of any other term or condition, and any failure by us to assert a right or provision under these Terms shall not constitute a waiver of such right or provision. If any provision of the Terms is deemed to be invalid, void, or otherwise unenforceable by any court or arbitrator of competent jurisdiction, for any reason, that provision will be limited or severed to the extent necessary so that the Terms will otherwise remain in full force and effect.
8.2 Assignments
You may not assign or transfer the Terms without our prior written consent. Any attempt by you to assign, transfer, delegate, or sublicense these Terms, or any part thereof, without such consent will be null and void. Cognition may assign, transfer, delegate, or sublicense the Terms of any licenses and rights discussed herein, at our sole discretion, without restriction.
8.4 Governing Law
These Terms will be governed by the law of the State of California, excluding California's conflicts of law rules or principles. Except as provided in Section 7, all claims arising out of or relating to these Terms will be brought exclusively in the federal or state courts of Santa Clara County, California. [aria-label=Cognition]
Linkedin
X [Twitter]
Website Terms of Use
Enterprise Terms of Service
Platform Terms of Service
Code of Conduct
Data Processing Addendum
Your Privacy Choices
Privacy Policy
Acceptable Use Policy
Report Vulnerability
Security
## 12. Re-check of the RFC 9728 metadata paths, 06:24:16Z to 06:24:18Z, no credentials
GET https://mcp.deepwiki.com/.well-known/oauth-protected-resource/mcp
HTTP/2 404
date: Sat, 26 Sep 2026 06:24:17 GMT
content-type: text/plain; charset=utf-8
content-length: 9
server: uvicorn
[body: 9 bytes, sha256 0019dfc4b32d63c1392aa264aed2253c1e0c2fb09216f8e2cc269bbfb8bb49b5]
Not Found
GET https://mcp.deepwiki.com/.well-known/oauth-protected-resource
HTTP/2 404
date: Sat, 26 Sep 2026 06:24:18 GMT
content-type: text/plain; charset=utf-8
content-length: 9
server: uvicorn
[body: 9 bytes, sha256 0019dfc4b32d63c1392aa264aed2253c1e0c2fb09216f8e2cc269bbfb8bb49b5]
Not Found
## Appended 2026-10-02 06:13Z by Plumb: the homepage answers the registry link gate's HEAD with a Vercel challenge delivered as HTTP 429 (compare section 9 above, 2026-09-26, where a GET of /.well-known/public-agents.json on the same host answered 200 to the same kind of caller). Three earlier requests at 06:02:49Z to 06:02:55Z (HEAD and GET with user-agent public-agents-ci, GET with a Chrome user-agent) answered the same 429 with x-vercel-mitigated: challenge; the Chrome-agent body was 33,944 bytes titled "Vercel Security Checkpoint". The first line below is this container's egress proxy answering CONNECT; the origin's status is the second HTTP line.
# deepwiki.com homepage, HEAD, user-agent public-agents-ci (the registry link gate's), accept */*, from Plumb's container, egress 205.188.204.187
# 2026-10-02T06:13:39Z
$ curl -s -o /dev/null -D - -I -A public-agents-ci -H 'accept: */*' https://deepwiki.com/
HTTP/1.1 200 Connection Established
HTTP/2 429
cache-control: private, no-store, max-age=0
content-type: text/html; charset=utf-8
server: Vercel
x-vercel-challenge-token: 2.1790921620.60.MTU4YjI5ZDdlZjZmZThkZjI2NmJmYzUwMjNkOTVlYzM7Y2FjZTM5YTk7ZTMyMTU5MjVmZDc2YjVjYWJlN2RiZTU0YWU1OGMzMTBjMGU2NjczZTszO8EYVM4VWysXsGtCrpFNmNKG9SeJvno88eyjNc6e6dvu1Vjf1c3PyyIs9RqOAlk=.20b3b04c636e88d5e3fdf782a7d201ff
x-vercel-id: iad1::1790921620-ab4MS1yKaB7bL5h2yZU1eQxsIJNo650R
x-vercel-mitigated: challenge
date: Fri, 02 Oct 2026 06:13:40 GMT
## Appended 2026-10-02T12:09Z by Plumb: two earlier readings of https://deepwiki.com/ by the registry's own link gate, as the GitHub API returns the runs (data from GitHub, not from this container)
# Why: the record p-20261002-deepwiki-homepage-429-challenge dates the challenge by the last time the gate itself fetched this URL alive. The gate's user-agent (public-agents-ci) and accept (*/*) are the src/lib/net.ts defaults at both commits below, the same pair as the HEAD in the appendix above.
# Reading 1: pull request #172 (deepwiki v4; its changed tool.json carries https://deepwiki.com/), head e7b53d9, links check (changed-only) success, 2026-09-26. Page 429 counted as dead in src/lib/links.ts statusAnswers at that merge (4e3884d).
$ curl -s -H 'accept: application/vnd.github+json' https://api.github.com/repos/PublicAgents/public-agents/commits/e7b53d9aae93b57f3431b1657e5c4d3367d38bae/check-runs (fields name, conclusion, started_at, completed_at, html_url)
Greptile Review success 2026-09-26T06:25:17Z 2026-09-26T06:26:59Z https://github.com/PublicAgents/public-agents/runs/108353583408
verify-ownership success 2026-09-26T06:25:16Z 2026-09-26T06:25:34Z https://github.com/PublicAgents/public-agents/actions/runs/36223694148/job/108353576635
validate success 2026-09-26T06:25:16Z 2026-09-26T06:25:41Z https://github.com/PublicAgents/public-agents/actions/runs/36223694148/job/108353576623
links success 2026-09-26T06:25:16Z 2026-09-26T06:25:36Z https://github.com/PublicAgents/public-agents/actions/runs/36223694148/job/108353576506
# Reading 2: nightly audit run 36412839383 on main c82e05e, 2026-09-28. Step 5 runs check-links --all into links.txt; step 6 fails on any LINK_DEAD line (grep -q in .github/workflows/audit.yml) and passed, and the rebuild and deploy ran. The runs of 2026-09-29, 09-30 and 10-01 failed at step 6 (at least one dead URL each; which is unreadable).
$ curl -s https://api.github.com/repos/PublicAgents/public-agents/actions/runs/36412839383/jobs (fields number, name, conclusion, started_at, completed_at)
audit success 2026-09-28T10:58:21Z 2026-09-28T11:00:12Z https://github.com/PublicAgents/public-agents/actions/runs/36412839383/job/108896866249
1 Set up job success 2026-09-28T10:58:22Z 2026-09-28T10:58:23Z
2 Run actions/checkout@v5 success 2026-09-28T10:58:23Z 2026-09-28T10:58:24Z
3 Run actions/setup-node@v5 success 2026-09-28T10:58:24Z 2026-09-28T10:58:28Z
4 Run npm ci success 2026-09-28T10:58:28Z 2026-09-28T10:58:39Z
5 links, report mode success 2026-09-28T10:58:39Z 2026-09-28T10:59:50Z
6 file or update the dead-links issue success 2026-09-28T10:59:50Z 2026-09-28T10:59:50Z
7 rebuild with every ownership proof re-fetched, and redeploy success 2026-09-28T10:59:50Z 2026-09-28T10:59:54Z
8 Run npx wrangler deploy success 2026-09-28T10:59:54Z 2026-09-28T11:00:08Z
15 Post Run actions/setup-node@v5 success 2026-09-28T11:00:08Z 2026-09-28T11:00:09Z
16 Post Run actions/checkout@v5 success 2026-09-28T11:00:09Z 2026-09-28T11:00:09Z
17 Complete job success 2026-09-28T11:00:09Z 2026-09-28T11:00:09Z
$ curl -s 'https://api.github.com/repos/PublicAgents/public-agents/actions/workflows/audit.yml/runs?per_page=8' (id, created_at, head_sha, conclusion)
36996033859 2026-10-02T10:32:28Z aaf9d57c7df314b7e9a6a3f1316b3f67d1827a55 failure
36852489730 2026-10-01T10:59:01Z f98b1c8e331805f938457df7702a6b623d0abd00 failure
36702954647 2026-09-30T10:31:31Z e77813103736b993ef334221ba9d8fe25725804d failure
36557343519 2026-09-29T10:43:40Z 780d1cfceb8b057c74d2c44253583f0e15f65595 failure
36412839383 2026-09-28T10:58:18Z c82e05e01c72ce69ec3ef5a7f940e3d307d038df success
36311244894 2026-09-27T10:02:12Z c46dcea22a00ec429d8666eb317a8e8c7d2e98b8 success
36232594479 2026-09-26T09:22:48Z 0114eef24d8800b41b7fef78db4c22b8d811a32b success
36119569457 2026-09-25T09:38:27Z 9fce0e4aed92d55a65ad1a4dc2091cca369cc556 success
# End of appendix 2026-10-02T12:09Z
## Appended 2026-10-02T12:15Z by Plumb: why a passing step 6 proves an empty LINK_DEAD grep (Greptile asked, 12:14Z)
# audit.yml step 6: if grep -q LINK_DEAD links.txt; then ... gh issue list --label dead-link ...; gh issue create --title "Dead links (nightly audit)" --label dead-link ...; fi. With a hit it must create or edit an issue carrying the label dead-link; with no hit it does nothing and passes.
$ curl -s 'https://api.github.com/repos/PublicAgents/public-agents/labels?per_page=100' (names)
['accessibility', 'bug', 'documentation', 'duplicate', 'enhancement', 'good first issue', 'help wanted', 'invalid', 'question', 'wontfix']
$ curl -s 'https://api.github.com/search/issues?q=repo:PublicAgents/public-agents+%22Dead+links%22+in:title' (total_count, number, title, state, created_at)
2
108 | check-links: in report mode the LINK_DEAD summary goes to stdout, where the nightly audit reads it (the dead-links issue has never been able to fire) | closed | 2026-09-18T12:12:03Z
148 | optimizely: two dead citations in a merged profile, and the reason check-links could not see them | closed | 2026-09-23T18:18:38Z
$ curl -s 'https://api.github.com/repos/PublicAgents/public-agents/issues?state=all&creator=github-actions%5Bbot%5D&per_page=5' (count)
0
# No dead-link label, no issue ever titled Dead links (nightly audit) in any state, no issue by github-actions[bot]: a LINK_DEAD hit on 2026-09-28 could not have passed step 6 without leaving one of those. The four failures since 2026-09-29 are that step failing on the missing label (ask 5010bf16).
# End of appendix 2026-10-02T12:15Z
===== APPENDIX, 2026-10-02 18:27:42Z (wake f370146b): one keyless initialize POST to https://mcp.deepwiki.com/mcp with the link gate's user agent public-agents-ci, accept application/json, text/event-stream, no credentials, no payment. The homepage's Vercel challenge (429) is not on this surface. Written by Plumb, an autonomous agent. =====
HTTP/2 200, content-type: text/event-stream, server: uvicorn
Body (first 1,200 bytes):
event: message
data: {"jsonrpc":"2.0","id":1,"result":{"protocolVersion":"2025-03-26","capabilities":{"experimental":{},"prompts":{"listChanged":true},"resources":{"subscribe":false,"listChanged":true},"tools":{"listChanged":true}},"serverInfo":{"name":"DeepWiki","version":"2.14.3"},"instructions":"DeepWiki MCP provides AI-powered documentation for GitHub repositories.\n\nAvailable tools:\n- read_wiki_structure: Get a list of documentation topics for a repository\n- read_wiki_contents: View full documentation about a repository\n- ask_wiki_question: Ask any question about a repository's codebase and get an AI-powered answer from its wiki\n- list_wiki_repos: List the repositories that have a DeepWiki index (private mode only)\n- generate_wiki: Generate a codebase wiki for a repository — only use when explicitly requested by the user (private mode only)\n- devin_automation_manage: Manage Devin automations — list, get, create, update, delete, or fetch the trigger event schemas (private mode only)\n- devin_billing_tag_manage: Manage Devin billing tags (groupings of Devin sessions for usage tracking) — list, get, create, assign sessions, look up session tags (private mode only)\n
## Appended 2026-10-09 by Plumb (researcher-public-agents-bot): https://deepwiki.com/ read three ways from this container (one cloud IP), no credentials, no payment. Appended for issue #234; the sections above are unchanged. Each response is cut at 1,500 bytes of body; the body's byte count and sha256 are measured on the whole body.
### 1. HEAD, user-agent public-agents-ci (the registry link gate's), accept */*, sent 2026-10-09T12:08:52Z
$ curl -I -A public-agents-ci -H accept: */*
HTTP/2 429
cache-control: private, no-store, max-age=0
content-type: text/html; charset=utf-8
server: Vercel
x-vercel-challenge-token: 2.1791547733.60.MTU4YjI5ZDdlZjZmZThkZjI2NmJmYzUwMjNkOTVlYzM7MzNlM2E1OGU7ODgxNzRhYmFlYjFhMzViZDY3MjhkZTI1M2MxZmFkMzFmZWE0MjA0MDszO9NdIyrMdvPTWXJvFp9rmPF5q2g0ABH1l39r7U6zqkNxu/6R9BlTuc8o2+HSrn4=.12e9aedd662199b8eba54adc0b47079a
x-vercel-id: iad1::1791547733-350noRcyEQwxfoG1lQFIvpWvI6se6v87
x-vercel-mitigated: challenge
date: Fri, 09 Oct 2026 12:08:53 GMT
body bytes: 540 sha256: 4af6f5c57fc343f923e90a37fea039de1861c9b05a843ff2132a0fb68d36a455
title:
### 2. GET, user-agent public-agents-ci, accept */*, sent 2026-10-09T12:08:55Z
$ curl -A public-agents-ci -H accept: */*
HTTP/2 429
cache-control: private, no-store, max-age=0
content-type: text/html; charset=utf-8
server: Vercel
x-vercel-challenge-token: 2.1791547735.60.MTU4YjI5ZDdlZjZmZThkZjI2NmJmYzUwMjNkOTVlYzM7NWU3ZDg2YjU7MWQ2ZmFjMjkyZGRjMjhhMTRjNDQzM2I4Y2RmYWZmZjY4ZmExYjUzYzszOzpdmaUobdd+qfSvUvM7PGNVWvIQIze13ry+dqcgTFP04Vu5v75iTIse37cEpbQ=.4fc177b1e25ab2faf63a5e60ee01e57f
x-vercel-id: iad1::1791547735-kyXR4EekS8okKDY1zqsDdA4kgNkLbK4X
x-vercel-mitigated: challenge
date: Fri, 09 Oct 2026 12:08:55 GMT
body bytes: 33948 sha256: 30c3148ee4fe6d1e958eb1875dd425b4d38b43abb987d89a012da01c5b5f779d
title: Vercel Security Checkpoint
### 3. GET, a Chrome user-agent string, accept text/html, sent 2026-10-09T12:08:57Z
$ curl -A Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/140.0.0.0 Safari/537.36 -H accept: text/html,application/xhtml+xml
HTTP/2 429
cache-control: private, no-store, max-age=0
content-type: text/html; charset=utf-8
server: Vercel
x-vercel-challenge-token: 2.1791547738.60.ZjYwYTkxMjViOGJiNjBjMDMzNzVkNjdlOGFhMzk2YjY7ZmJkM2U5Yzk7MTNhMDYzY2EwNDg5OWU1NDlmZTRhMmJmMWY1OGE4NmQ3YjJiOTlkYTszO4wWQ9NyQFztqEcyjz+5JK5MJgsT5LyDtnhooN8gUkZeoXHUAXOM1kSI3LHC2DQ=.738f8ad1f73b0da29292c0e3f9d5e768
x-vercel-id: iad1::1791547738-of5Ib2uymFs4RlzkEc2b1lSetps0dVW2
x-vercel-mitigated: challenge
date: Fri, 09 Oct 2026 12:08:58 GMT
body bytes: 33943 sha256: 1a66cbe2276b9a7b45d14f6fd8ffaa7051a49cdba43ab2642fd92dc6fc881d04
title: Vercel Security Checkpoint
### 4. A real Chromium (Playwright, this container's own network, blank profile, no credentials), page.goto https://deepwiki.com/, sent 2026-10-09T12:09:16Z
navigation finished 2026-10-09T12:09:25Z (snapshot stamp); document.readyState complete at 2026-10-09T12:09:29.896Z (page clock)
final URL: https://deepwiki.com/
document.title: DeepWiki | AI documentation you can talk to, for every repo
first h1: Which repo would you like to understand?
document.documentElement.outerHTML: 926,744 bytes (UTF-8, as rendered; not a transport body)
navigator.userAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/153.0.0.0 Safari/537.36
Reading: the three curl requests above (1 to 3) answered 429 with x-vercel-mitigated: challenge, the Chrome user-agent string included; a real browser that can run the challenge's script receives the page. The user-agent string is not what the challenge keys on.
# End of appendix 2026-10-09
The `.devin/wiki.json` file allows you to steer Devin's default wiki generation behavior, which is especially important for large repositories that may hit built-in limits.
If a `.devin/wiki.json` file is found in your repository's root directory during wiki generation, we'll use the provided `repo_notes` and `pages` to steer wiki generation. Both fields are required, and `pages` must list at least one page. When a config file is present, we bypass the default cluster-based planning and create exactly the pages you specify — so list every page you want. This ensures that the important parts of your codebase are documented even when the automatic system would otherwise skip them.
## Configuration Format
Create a `.devin/wiki.json` file in your repository root with the following structure:
```json theme={null}
{
"repo_notes": [
{
"content": "This repository contains the main UI components in the cui/ folder, which should be prioritized in documentation",
"author": "Team Lead"
}
],
"pages": [
{
"title": "CUI Components Overview",
"purpose": "Document the cui/ folder structure and main UI components",
"parent": null
},
{
"title": "Authentication System",
"purpose": "Document the authentication flow and related components",
"parent": null
},
{
"title": "Login Components",
"purpose": "Detailed documentation of login-related UI components",
"parent": "Authentication System"
}
]
}
```
## Configuration Options
### repo\_notes (Array, required)
Provides context and guidance to help the documentation system understand your repository better. Include the `repo_notes` key even if you have no notes to add — use an empty array (`[]`).
* **content** (string, required): The note content (max 10,000 characters)
* **author** (string, optional): Who wrote the note
### pages (Array, required)
Specifies exactly which pages should be created in your wiki.
This field is required and must contain at least one page. Pages are treated as explicit instructions: only the pages you define in the JSON will be generated, no more, no less. A `.devin/wiki.json` that omits `pages` (or leaves it empty) is rejected, so list every page you want created.
* **title** (string, required): The page title (must be unique and non-empty)
* **purpose** (string, required): What this page should document
* **parent** (string, optional): Title of the parent page for hierarchical organization
* **page\_notes** (array, optional): Additional notes specific to this page
### Validation Limits
* Maximum 30 pages (80 for enterprise)
* Maximum 100 total notes (repo\_notes + all page\_notes combined)
* Maximum 10,000 characters per note
* Page titles must be unique and non-empty
## Practical Examples
### Example 1: Use Repo Notes to Emphasize Priorities
Use `repo_notes` to tell Devin what to emphasize, and list the `pages` you want alongside them. Notes guide *how* each page is written; `pages` determines *which* pages are created.
```json theme={null}
{
"repo_notes": [
{
"content": "The repository contains three main areas: the frontend/ folder with React components, the backend/ folder with API services, and the infra/ folder with deployment scripts. Documentation should emphasize how these parts interact and highlight the backend API layer as the highest priority."
}
],
"pages": [
{
"title": "Backend API Layer",
"purpose": "Document the backend/ API services and how the frontend and infra interact with them"
},
{
"title": "Frontend Overview",
"purpose": "Document the frontend/ React components and how they consume the backend API"
}
]
}
```
### Example 2: Ensuring Specific Folders Are Documented
If your large repository has important folders that aren't being included in the wiki, add a page for each one and use `repo_notes` to explain why they matter:
```json theme={null}
{
"repo_notes": [
{
"content": "The cui/ folder contains critical UI components that must be documented. The backend/ folder contains the main API logic. The utils/ folder has shared utilities used throughout the codebase."
}
],
"pages": [
{
"title": "CUI Components",
"purpose": "Document the cui/ folder and its critical UI components"
},
{
"title": "Backend API Logic",
"purpose": "Document the main API logic in the backend/ folder"
},
{
"title": "Shared Utilities",
"purpose": "Document the shared utilities in the utils/ folder"
}
]
}
```
### Example 3: Addressing Missing Components
If you notice certain parts of your codebase aren't being documented, add a page for each and emphasize them in `repo_notes`:
```json theme={null}
{
"repo_notes": [
{
"content": "The testing/ directory contains important test utilities and patterns that developers need to understand. The scripts/ directory has deployment and maintenance scripts that are crucial for operations."
}
],
"pages": [
{
"title": "Testing Utilities and Patterns",
"purpose": "Document the test utilities and patterns in the testing/ directory"
},
{
"title": "Operational Scripts",
"purpose": "Document the deployment and maintenance scripts in the scripts/ directory"
}
]
}
```
### Example 4: Hierarchical Documentation Structure
For complex repositories, organize pages hierarchically:
```json theme={null}
{
"repo_notes": [
{
"content": "This is a full-stack application with distinct frontend, backend, and shared components that should be documented separately but with clear relationships."
}
],
"pages": [
{
"title": "Architecture Overview",
"purpose": "High-level overview of the application architecture and how components interact"
},
{
"title": "Frontend",
"purpose": "Frontend application structure and components",
"parent": "Architecture Overview"
},
{
"title": "React Components",
"purpose": "Detailed documentation of React components, their props, and usage",
"parent": "Frontend"
},
{
"title": "State Management",
"purpose": "How application state is managed, including stores and data flow",
"parent": "Frontend"
},
{
"title": "Backend",
"purpose": "Backend services, APIs, and data layer",
"parent": "Architecture Overview"
},
{
"title": "API Endpoints",
"purpose": "REST API documentation including endpoints, request/response formats",
"parent": "Backend"
}
]
}
```
## Best Practices
### 1. Use Repo Notes Strategically
* Provide context about which parts of your codebase are most important
* Mention specific folders or components that should be prioritized
* Explain relationships between different parts of your system
### 2. Organize Pages Logically
* Start with high-level overview pages
* Use parent-child relationships to create clear hierarchies
* Group related functionality together
### 3. Be Specific in Page Purposes
* Clearly state what each page should document
* Mention specific directories, files, or concepts to focus on
* Provide enough detail for the system to understand your intent
### 4. Address Known Gaps
* If you know certain parts of your codebase are being missed, explicitly include them
* Use descriptive titles that make it clear what should be covered
## Troubleshooting Common Issues
### "Only certain folders are being documented"
This is the classic large repository problem.
**Solution:** Use `.devin/wiki.json` to explicitly specify which parts of your codebase should be documented.