← back NO. #7c26d4da

NO.: #7c26d4da
Topic: AGENTS & TOOLS
Source: Claude Code Releases
Published: 2026-04-23 00:42:21
Importance: ★ 7/10 — radar 70
Original: github.com/anthropics/claude-code/releases/tag/v2.1.118

`Claude Code` `v2.1.118` adds visual mode, `/usage`, and direct MCP hooks

Editing and ops got sharper at once: Vim visual mode landed, theme/plugin workflows expanded, and hooks can call MCP tools directly. This is a practical quality jump if you script Claude Code heavily or run it across managed environments.

[ KEY POINTS ]

Vim v and V visual modes now support selection, operators, and visual feedback, cutting friction for keyboard-first editing inside the CLI.
/cost and /stats are folded into /usage, while old commands still work as shortcuts, so telemetry gets simpler without breaking muscle memory.
Hooks can invoke MCP tools via type: "mcp_tool", which opens cleaner automation paths than shell-only hook chains.
DISABLE_UPDATES blocks all update paths, and WSL can inherit Windows-managed settings, making locked-down team setups easier to enforce.
A long list of MCP OAuth and credential fixes targets re-auth churn, token races, and broken auth states, so integrations should fail less often.

Originalgithub.com/anthropics/claude-code/releases/tag/v2.1.118Read original →

// related

#0001
#0001Agents & tools Google AI Forum1 hour ago
`Antigravity 2.0` quota complaints: `Flash` loses its lightweight edge
40radar
AntigravityAI coding agent — Google-model-based dev workflow
A tiny one-change request reportedly burned about 50 credits, making 1,000 monthly credits thin for daily work. Quota design now matters as much as model quality.
- Flash and Pro now effectively draw from the same quota pool, removing the cheaper lane for iterative edits.
- Partial refresh moved from roughly 5 hours into a 7-day wait pattern after recovery, weakening reliability for long coding sessions.
- At ~50 credits for a small change, the monthly allocation supports only limited agent work before tool switching becomes rational.
Source: discuss.ai.google.dev/t/ai-model-quota-limits-have-no-usRead original →
FIG-0011:1
40radar
FIG-0011:1
#0002
#0002Agents & tools Google AI Forum3 hours ago
`Antigravity` `v2.0.0` Breakage: Staging Endpoint Misconfigured
50radar
AntigravityAI coding IDE — agent-based development workflows
Chat, completions, and agents can fail because the release points at a staging Cloud Code endpoint. Manual v2.0.1 reinstall is the practical fix when auto-update stalls.
- language_server.exe launches with --cloud_code_endpoint https://daily-cloudcode-pa.googleapis.com; production OAuth tokens get rejected there.
- Failures repeat every 25-30 seconds, breaking loadCodeAssist, fetchUserInfo, and fetchAvailableModels downstream.
- v2.0.1 reportedly clears the daily- endpoint, but some auto-updaters download to %LOCALAPPDATA%\antigravity-updater\pending\ without applying.
- Windows users stuck on v2.0.0 should skip debugging prompts and reinstall manually; this is a release config bug, not a project setup issue.
Source: discuss.ai.google.dev/t/fix-for-antigravity-v2-0-0-issueRead original →
FIG-0021:1
50radar
FIG-0021:1
#0003
#0003Agents & tools Google AI Forum3 hours ago
`Antigravity` users push back on hidden compute quotas
60radar
AntigravityAI coding IDE — agent workflow powered by Google models
Paid usage turned unpredictable after quota accounting moved from requests to hidden compute. A cheaper plan can now fail mid-workflow; budget risk matters more than model choice.
- A $20 Pro user reports 2-3 weeks of HTTP 429 lockouts despite visible quota remaining; reliability became the real blocker.
- The May 19 change replaced request limits with hidden compute-used accounting, making agent background scans and micro-queries harder to budget.
- Gemini 3.5 Flash is described as more verbose and weaker at coding, burning quota through long explanations instead of useful edits.
- Upgrade pressure jumps to $100-$200 Ultra or a reported 5-day ban after quota exhaustion; keep fallback IDE and agent paths ready.
Source: discuss.ai.google.dev/t/how-antigravity-became-cursor-2-Read original →
FIG-0031:1
60radar
FIG-0031:1

`Claude Code` `v2.1.118` adds visual mode, `/usage`, and direct MCP hooks

// related

`Antigravity 2.0` quota complaints: `Flash` loses its lightweight edge

`Antigravity` `v2.0.0` Breakage: Staging Endpoint Misconfigured

`Antigravity` users push back on hidden compute quotas