This release adds 2 notable features for engineering teams evaluating rollout.
Published 28d
AI Agents & Assistants
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
agent
cargo
cli
codingagent
crossterm
ratatui
+2 more
rust
tui
Summary
AI summaryBroad release touches Other, Bug Fixes, @kernitus, and Tests.
Full changelog
What's Changed
0.133.22 - 2026-06-28
Highlights
Bug Fixes
- Stream isolated review subcalls (a7ffd7f7) (@kernitus)
- Ignore responses keepalive events (f33650f5) (@kernitus)
- Update compatible responses retry test (7b73ba2c) (@kernitus)
Documentation
- Add web_search tool reference, document /continue and did-you-mean (90a321b2)
- Expand TODO with known tech debt items (144491a0)
- Add defuddle_fetch tool reference and web_search config fields (5040faca)
- Fix stale src/tui.rs references, update CONTRIBUTING.md and INDEX.md (8b0518e1)
- Audit and trim per-crate AGENTS.md files (018656ea)
- Align CI/CD docs with actual workflow files (5e39c041)
Features
- Implement cross-turn tracking and limit assistant text responses to prevent runaway loops (466c0c24)
- Add tiktoken integration for token counting and hierarchical summarization (3d5af4ed)
Other Changes
Other
- Convert stringly-typed ResponseId, OutputItemId, ContentPartId to newtypes (2df60016)
- Implement recovery fallback messages and enhance outline search with directory summaries (75429ee7)
- Implement few-shot example selection and rendering; enhance tool descriptions and documentation. (82f078bb)
- Add persistent audit logging and per-MCP-server sandboxing (086208e3)
- Implement async approval blocker and escalation gate for irreversible actions (32d15be6)
- Add task history, message metadata, pending actions, and state schema modules (b670e3a0)
- Enhance tool descriptions for clarity and guidance on usage (b071d6a5)
- Enhance agent execution flow with escalation handling and stagnation detection (209364a4)
- Optimize token count management and enhance loop detection efficiency (e1f90cec)
- Refactor escalation decision enum and clean up evaluator gate outcome handling (c27527db)
- Refine system prompt guidelines for clarity and simplicity in agent operations (ae9a8a81)
- Clarify description for 'follow' parameter in unified search schema (e86300cc)
- Enhance skill availability messages to include instructions for listing full catalog (5512d9df)
- Refactor tool outcome handling and improve skill overflow messages for clarity (782b8fc7)
- Refactor model ID matching and clean up import statements for clarity (764128be)
- Refactor string formatting for improved readability and consistency across tests and core modules (87a375f3)
Tests
- Fix prompt cache retention fixtures (6d032016) (@kernitus)
Full Changelog: https://github.com/vinhnx/vtcode/compare/0.133.21...0.133.22
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About VT Code
All releases →Related context
Related tools
Beta — feedback welcome: [email protected]