Trajectory ID: 29f0c19d-1627-492a-8d90-c10aa986401bz
The Issue: A persistent “agent executor error” occurs, even when sending a simple message like “hi”. The error states: “could not convert a single message before hitting truncation.”
Root Cause: The system has 300 installed skills. While other tools use lazy loading and work without issue, Antigravity’s TrajectoryChatConverter appears to load all skill metadata, causing the first message to exceed the token limit.
Stack Trace Snippet: google3/third_party/jetski/cortex/chatconverters/chatconverters.(*TrajectoryChatConverter).TrajectoryToChatMessagesFromNCheckpointsAgo
Request: Implement a search-based or lazy-loading mechanism for skills in the Cortex/Cascade manager to prevent session crashes with large skill libraries.
hi i got the same problem
Trajectory ID: fe8e6650-3bb2-479
Error: agent executor error: could not convert a single message before hitting truncation
(1) attached stack trace
– stack trace:
| google3/third_party/jetski/cortex/cortex.(*CascadeManager).executeHelper.func1
| third_party/jetski/cortex/cascade_manager.go:2088
| […repeated from below…]
Wraps: (2) agent executor error
Wraps: (3) attached stack trace
– stack trace:
| google3/third_party/jetski/cortex/chatconverters/chatconverters.init
| third_party/jetski/cortex/chatconverters/trajectory_chat_converter.go:44
| runtime.doInit1
| third_party/go/gc/src/runtime/proc.go:8105
| runtime.doInit
| third_party/go/gc/src/runtime/proc.go:8072
| runtime.main
| third_party/go/gc/src/runtime/proc.go:258
| runtime.goexit
| third_party/go/gc/src/runtime/asm_amd64.s:1774
Wraps: (4) could not convert a single message before hitting truncation
Error types: (1) *withstack.withStack (2) *errutil.withPrefix (3) *withstack.withStack (4) *errutil.leafError
I’ve found a solution for this issue. The error seems to be caused by conflicting files in the “skills” directory.
I recommend backing up the files in the following path and then deleting them: C:\Users\.gemini\antigravity\skills
Once cleared, try running it again with any model, it should work perfectly 
Yes, after clearing , it worked perfectly but the issue is we cant use the global skills in gemini/antigravity like others .
Hello @Divith_S @epicvoxel, welcome to AI Forum!
Thank you for your insightful feedback on the issue while using large number of global skills. We appreciate you taking the time to share your thoughts with us, and we’ll be filing a feature request.
To help us prioritize this request effectively, any additional details, screenshots, error logs etc. you can provide would be very helpful.
I found a workaround that’s been working for me.
It requires 2+ Google accounts (ideally both with Pro/Ultra access). You can switch between them using Antigravity Manager: https://github.com/lbjlaq/Antigravity-Manager
If you hit the errors I mentioned earlier (or the one @Divith_S reported), try switching accounts once. In most cases that resets things and the agent runs again. If it still doesn’t work after the first switch, switch again — sometimes it takes a couple of tries.
Not a proper fix, but a reliable temporary workaround until this gets resolved.