Connecting agents: questions and answers
Questions about connecting Claude Code, Codex, Cursor and other MCP clients to Talewright: browser sign-in and consent, connected apps, choosing between an API key and browser sign-in, and what an agent can and cannot do for a member. The rules are in Connected apps and Credentials and scopes.
Choosing how to connect
- How do I connect an AI agent to Talewright?
- What is a connected app?
- What is the difference between a connected app and an API key?
- Should I use an API key or browser sign-in for my agent?
- Which apps can sign in through the browser?
- Can I connect an app other than Claude or Codex through the browser?
- Do the same rules apply to a connected app and an API key?
- Do I need to be an owner or admin to connect an agent?
- Do I need a Talewright account before I connect an agent?
- Does the agent ever see my password?
- Does a connected app work with the REST API?
- Why does my connected app's token work at
/mcpbut not at/v1? - Does my agent need my browser to stay open?
- Is there anything to install to use Talewright's MCP server?
- Where is the guide for agents that Talewright serves?
Connecting Claude Code
- How do I connect Claude Code to Talewright?
- Where does the app show the command for Claude Code?
- How do I sign in to Talewright from inside Claude Code?
- Why is Talewright missing when I open Claude Code in another project?
- How do I connect Claude Code with an API key instead of browser sign-in?
- What should I do in Claude Code after I revoke the API key it uses?
- Can I connect Claude Code on two computers to the same organization?
- Does signing in from Claude Code on a second computer end the connection on the first?
- Why does Claude Code say it is not authorized for Talewright?
Connecting Codex
Connecting Cursor and other MCP clients
- How do I connect Cursor to Talewright?
- Can Cursor sign in to Talewright through the browser?
- How do I connect an MCP client that is not listed?
- Which scopes should the API key of an MCP client have?
- Can an agent that runs in a web page connect to Talewright?
- Can an app register itself with Talewright to connect?
The consent screen
- What happens in my browser when an app asks to connect?
- What does the consent screen tell me about the app?
- What do I choose on the consent screen?
- Can I give a connecting app less than it asked for?
- Can I give a connecting app a permission it did not ask for?
- Why are only some permissions shown on the consent screen?
- Which permissions does an app get when it asks for none?
- Which permissions are ticked when the consent screen opens?
- Why does "Set file statuses" start unticked although the app asked for it?
- Why did "Set file statuses" untick itself on the consent screen?
- Who can grant a connecting app
files:approve? - Why is a workspace already ticked on the consent screen?
- Can I connect an app to two organizations in one step?
- What happens when I press "Allow"?
- What happens when I press "Deny"?
- I pressed "Deny" by mistake. How do I connect the app after all?
- What happens if I press "Allow" twice or open the request in two tabs?
- Can I correct my choice when "Allow" is refused?
- Can I allow an app with no permission ticked?
- I was signed out when the app opened my browser. Is the request lost?
- Can I connect an app when my account reaches no workspace?
- Can someone else answer the request my app opened?
- What if someone else opens my consent link before I do?
- I opened the request signed in to the wrong account. Can I move it to the right one?
- Why does a request to connect stay with the account that opened it first?
- Who owns a request that is decided before anyone has read it?
- Can an agent approve its own request to connect?
- Where do I go after I decide on the consent screen?
- When exactly does an app become connected?
- Is my access checked again after I press "Allow"?
- Is an email sent when an app is connected?
Lifetimes and tokens
- How long does a connection to an app last?
- Do I have to approve the app again every hour when its access token expires?
- Why did my agent lose access after a month without use?
- What happens when the lifetime of a connection ends?
- Can I extend a connection without connecting again?
- How do I change the workspaces or permissions of a connected app?
- Can a connected app give itself more permissions when it renews its token?
- What are the
at1_andrt1_tokens? - What happens if a one-time code or a refresh token is stolen and used again?
- Can one app use another app's tokens?
- Can a refresh token be sent as a bearer token?
Several connections per app
- Can I connect the same app twice for one organization?
- How many connections to one app can I have in one organization?
- What happens when I allow the same app an eleventh time for one organization?
- How do I tell two connections of the same app apart?
- Does disconnecting one connection end my other connections to the same app?
- Can I connect the same app to two organizations?
- Can I connect both Claude Code and Codex to the same organization?
- What is recorded when I connect an app again?
Listing and disconnecting
- Where do I see my connected apps?
- What does the list show for each connected app?
- How do I disconnect an app?
- What happens when I disconnect an app?
- Can I undo disconnecting an app?
- What happens to the work an app did after I disconnect it?
- What does the dialog "Disconnect {name}?" say?
- In which ways can a connection to an app end?
- Can an app disconnect itself?
- Can an owner or admin disconnect my connected app?
- What happens to a running agent when I disconnect it?
- How do owners and admins see the apps members connected?
- How do I see what a connected app did?
- What is the "last client" shown for a connected app?
- Which operations are behind the connected apps page?
What an agent can and cannot do for you
- What can an agent do once it is connected?
- Does an agent act as me or as itself?
- Can an agent do more than I can do myself?
- Can an agent find and read my files?
- Can an agent upload files for me?
- Can an agent rename, move, tag and describe files?
- Can an agent restore something from the trash?
- Can an agent set a file's status?
- Can an agent make a file public?
- Can an agent delete a file for good?
- Can an agent prepare changes for me to review first?
- Can an agent apply the suggestions it made?
- Can an agent connect another agent or create an API key?
- Can an agent see my other connected apps and API keys?
- Can an agent work in my other organizations?
- What happens to my agent when I am removed from one of its workspaces?
- What happens when I am demoted while my app holds
files:approve? - How do I limit what an agent can do?
- How do I stop an agent at once?
- How do I see what an agent did?
- Can an agent hide what it did?
- Is an agent's proposal still there after I disconnect the agent?
- Who is recorded when I apply an agent's proposal?
For people who build an MCP client
- Which OAuth flow does Talewright use for connected apps?
- Which OAuth endpoints does Talewright have?
- How does an MCP client find out how to sign in?
- What does the protected resource document say?
- What does the authorization server document say?
- Does Talewright support OpenID Connect?
- Does Talewright support the client credentials or device code grant?
- How does an app identify itself to Talewright?
- What must an app's client metadata document look like?
- Do connecting apps have a client secret?
- Which name does the consent screen show for an app?
- How often does Talewright read an app's metadata document?
- Which redirect addresses may an app use?
- Can a command-line app use a different loopback port each time?
- Which parameters does
GET /oauth/authorizetake? - Is PKCE required to connect an app?
- What must the
resourceparameter be? - What does
GET /oauth/authorizeanswer? - What does an app send to exchange its code for tokens?
- What does an app send to refresh its tokens?
- What does the token endpoint answer?
- Does each refresh give a new refresh token?
- Is the one-time code used up when the exchange fails?
- What happens when a refresh asks for fewer scopes?
- How does an app revoke its own connection?
- Can one app revoke another app's token?
- Do the OAuth endpoints use my browser session?
- Are the OAuth endpoints in the OpenAPI contract?
- What does a connected app read about itself?
- What does the decision on the consent screen send?