Skip to main content
ollim-bot controls which tools the agent can use through a permission system built on top of the Claude Agent SDK’s canUseTool callback. You choose a permission mode, and the bot enforces it — from silent denial of unapproved tools all the way to full bypass.

​
Permission modes

Switch modes with the /permissions slash command. The mode is fork-scoped: changing it inside an interactive fork does not affect the main session, and vice versa.
ModeBehavior
dontAskSilent denial of tools not in the session-allowed set. No prompt. Default.
defaultDiscord message with reaction buttons for each tool use.
acceptEditsSDK built-in — file-edit tools auto-approved, others prompt.
bypassPermissionsSDK built-in — all tools auto-approved, no prompts.
dontAsk is ollim-bot’s own layer. Under the hood, the SDK permission mode stays at default — the bot’s canUseTool callback handles the silent denial before the SDK ever sees it.

​
Approval flow

In default mode, unapproved tool use triggers an interactive approval prompt in your Discord DM. (acceptEdits auto-approves file-edit tools at the SDK level, and bypassPermissions approves everything at the SDK level — neither mode triggers the interactive flow below.)
1

Agent requests a tool

The SDK invokes the canUseTool callback with the tool name and input data.
2

Bot sends an approval message

A message appears in your DM showing the tool label (name + arguments). Three emoji reactions are added automatically.
3

You react

React with one of the three emojis to respond:
ReactionEffect
✅Allow — approve this single tool use
❌Deny — reject this tool use
🔓Always — approve and add the tool to the session-allowed set
4

Result is applied

The approval message is edited to show the outcome (allowed, denied, or always allowed), and the agent proceeds or receives a denial.
Approvals time out after 60 seconds. An unanswered prompt is treated as a denial, and the message is edited to show timed out.

​
Session-allowed set

When you react with 🔓 (Always), the tool is added to a session-allowed set. Future uses of that tool are auto-approved without prompting — even in dontAsk mode.
  • The session-allowed set is shared across the main session and interactive forks.
  • /clear resets the session-allowed set (and cancels any pending approvals).
  • The set is in-memory only — it does not persist across bot restarts.

​
Background forks

Background forks always deny tool permission requests, regardless of the current mode. The agent in a background fork cannot trigger approval prompts. Tools available to background forks are controlled separately through allowed_tools and disallowed_tools in the fork’s YAML configuration — see Background forks for details.

​
Cancellation

Pending approval prompts are automatically cancelled (treated as denials) when any of the following occur:
  • You use /clear — resets the session-allowed set and cancels all pending approvals
  • You interrupt the agent — cancels pending approvals for the current turn
  • You exit an interactive fork — cancels any approvals pending inside the fork
Cancelled prompts are edited to show cancelled with strikethrough text.

​
Examples

In dontAsk mode, any tool not in the session-allowed set is silently denied — no message appears.
/permissions dontAsk
This is the default. The agent operates autonomously within its allowed toolset without interrupting you for approvals.

​
Next steps

Slash commands

Full reference for all slash commands including /permissions.

Forks

How interactive and background forks interact with permission scoping.

MCP tools

Reference for all MCP tools the agent can use.

Background forks

Tool restrictions and allowed/disallowed tool configuration for background forks.