Feature/vscode extension#4

Closed

KillerQueen-Z wants to merge 185 commits intoBlockRunAI:mainfrom

KillerQueen-Z:feature/vscode-extension

Commits on Mar 22, 2026

Commits on Mar 23, 2026

refactor: use @blockrun/llm for wallet management and x402 signing
1bcMax
committed
feat: publish-ready — BUSL license, battle-tested README, npm package config
1bcMax
committed
chore: use @blockrun/llm 1.4.2 from npm
1bcMax
committed
chore: rename to @blockrun/cc for npm
1bcMax
committed
docs: update install command to @blockrun/cc
1bcMax
committed
feat: dual chain support — brcc setup base|solana
1bcMax
committed
chore: bump 0.3.0 — dual chain support
1bcMax
committed
fix: avoid auth conflict — clean oauth tokens, use realistic API key format
1bcMax
committed
feat: add --model flag and brcc models command — list 40+ models with pricing
1bcMax
committed
fix: remove unsupported --api-key-auth flag
1bcMax
committed
feat: proxy model override — brcc start --model routes all requests to chosen model
1bcMax
committed
docs: add ROADMAP — tweakcc + CCR integration plan, phases 1-4
1bcMax
committed
feat: Phase 1 — env var model mapping, config command, smart routing infrastructure
1bcMax
committed
feat: Phase 2 — install script, config command, models command, dual chain docs, README overhaul
1bcMax
committed
docs: 10K-star README — badges, comparison table, architecture diagram, issue links
1bcMax
committed
docs: remove unnecessary legal FAQ
1bcMax
committed
docs: trim FAQ
1bcMax
committed
docs: add launch blog — Claude Code Rate Limits Are Broken. Here's the Fix.
1bcMax
committed
docs: add sudo for Linux npm install
1bcMax
committed
fix: cap max_tokens for models with lower limits (DeepSeek 8192, Haiku 8192)
1bcMax
committed
chore: bump 0.6.1
1bcMax
committed
fix: cap max_tokens to 16384 default — reduces per-call cost from ~$0.50 to ~$0.05
1bcMax
committed
feat: adaptive max_tokens — use last output_tokens for pricing, first call uses 4096 default
1bcMax
committed
fix: debug logging behind --debug flag, silent by default
1bcMax
committed
fix: adaptive max_tokens uses max(lastOutput*2, 4096) — prevents starvation; debug logs to stderr
1bcMax
committed
fix: debug logs to file (~/.blockrun/brcc-debug.log) instead of stderr
1bcMax
committed
feat: in-session model switching — type 'use gpt' or 'use deepseek' inside Claude Code
1bcMax
committed
fix: handle array content format in model switch detection, add debug for message inspection
1bcMax
committed
fix: add request-level debug logging to trace model switch persistence
1bcMax
committed

Commits on Mar 24, 2026

Commits on Mar 26, 2026

Commits on Mar 30, 2026

chore: track dist in git, remove from .gitignore
1bcMax
committed

Commits on Mar 31, 2026

feat: init/daemon/uninit commands, smart routing default, fix OAuth vars, restore sse-translator.ts, bump v0.9.13
1bcMax
committed

Commits on Apr 2, 2026

Commits on Apr 3, 2026

fix: useInput/TextInput conflict resolved, ImageGen 60s timeout, clean exit
1bcMax
committed

Commits on Apr 4, 2026

Commits on Apr 5, 2026

Commits on Apr 6, 2026

Commits on Apr 7, 2026

Commits on Apr 8, 2026

Add /history and /delete slash commands
KillerQueen-Z
committed

Commits on Apr 9, 2026

Enhance VS Code extension support and model management
KillerQueen-Z
committed