marketing-shibata50/claude-code-ultimate-guide
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

release: v3.27.6 - Sonnet 4.6 default + 200K vs 1M context guide

Browse source 
- Pricing table: Sonnet 4.6 now default (Feb 2026)
- New section: 200K vs 1M context decision guide (MRCR bench, cost table, use cases)
- threat-db.yaml v2.1.0: CVE-2026-23744, Slopsquatting T009, OWASP Agentic AI Top 10

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Florian BRUNIAUX 2026-02-18 09:33:55 +01:00
parent 4b15bdb137
commit c3da456d3a
8 changed files with 173 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

4
guide/cheatsheet.md
View file

@ -12,7 +12,7 @@ tags: [cheatsheet, reference]
**Written with**: Claude (Anthropic)
**Version**: 3.27.5 | **Last Updated**: February 2026
**Version**: 3.27.6 | **Last Updated**: February 2026
---
@ -544,4 +544,4 @@ where.exe claude; claude doctor; claude mcp list
**Author**: Florian BRUNIAUX | [@Méthode Aristote](https://methode-aristote.fr) | Written with Claude
*Last updated: February 2026 | Version 3.27.5*
*Last updated: February 2026 | Version 3.27.6*

49
guide/ultimate-guide.md
View file

@ -16,7 +16,7 @@ tags: [guide, reference, workflows, agents, hooks, mcp, security]
**Last updated**: January 2026
**Version**: 3.27.5
**Version**: 3.27.6
---
@ -1738,11 +1738,12 @@ Claude Code isn't free - you're using API credits. Understanding costs helps opt
#### Pricing Model (as of February 2026)
Claude Code uses **Claude Sonnet 4.5** by default:
Claude Code uses **Claude Sonnet 4.6** by default (as of Feb 2026):
| Model | Input (per 1M tokens) | Output (per 1M tokens) | Context Window | Notes |
|-------|----------------------|------------------------|----------------|-------|
| **Sonnet 4.5** | $3.00 | $15.00 | 200K tokens | Default model |
| **Sonnet 4.6** | $3.00 | $15.00 | 200K tokens | Default model (Feb 2026) |
| Sonnet 4.5 | $3.00 | $15.00 | 200K tokens | Legacy (same price) |
| Opus 4.6 (standard) | $5.00 | $25.00 | 200K tokens | Released Feb 2026 |
| Opus 4.6 (1M context beta) | $10.00 | $37.50 | 1M tokens | Requests >200K context |
| Opus 4.6 (fast mode) | $30.00 | $150.00 | 200K tokens | 2.5x faster, 6x price |
@ -1750,6 +1751,42 @@ Claude Code uses **Claude Sonnet 4.5** by default:
**Reality check**: A typical 1-hour session costs **$0.10 - $0.50** depending on usage patterns.
#### 200K vs 1M Context: Performance, Cost & Use Cases
The 1M context window (beta, API only) is a significant capability jump — but it's not always the right choice.
**Retrieval accuracy at scale (MRCR v2 8-needle benchmark)**
| Model | 256K accuracy | 1M accuracy | Source |
|-------|--------------|-------------|--------|
| Opus 4.6 | 93% | 76% | Anthropic blog (Feb 2026) |
| Sonnet 4.5 | — | 18.5% | Anthropic blog |
| Sonnet 4.6 | Not yet published | Not yet published | — |
Note: Opus 4.6 retains strong accuracy at 1M (76%), while Sonnet 4.5 degrades sharply. Sonnet 4.6 MRCR scores have not yet been published by Anthropic.
**Cost per session (approximate)**
| Session type | ~Tokens in | ~Tokens out | Sonnet 4.6 | Opus 4.6 |
|---|---|---|---|---|
| Bug fix / PR review | 50K | 5K | ~$0.23 | ~$0.38 |
| Module refactoring | 150K | 20K | ~$0.75 | ~$1.25 |
| Full service analysis (1M beta) | 500K | 50K | ~$2.25 | ~$8.75 |
**When to use which**
| Scenario | Recommendation |
|----------|---------------|
| Bug fix, PR review, daily coding | Sonnet 4.6 @ 200K — fast and cheap |
| Cross-module refactoring, large codebase | Sonnet 4.6 @ 1M beta — volume without premium |
| Architecture analysis, Agent Teams, complex reasoning | Opus 4.6 @ 1M beta — accuracy matters |
**Key facts**
- Opus 4.6 max output: **128K tokens**; Sonnet 4.6 max output: **64K tokens**
- 1M context ≈ 30,000 lines of code / 750,000 words
- 1M context is **beta** (API only, requires `anthropic-beta: interleaved-thinking-2025-05-14` header)
- Opus 4.6 pricing doubles above 200K; no premium tier announced yet for Sonnet 4.6
#### What Costs the Most?
| Action | Tokens Consumed | Estimated Cost |
@ -3632,7 +3669,7 @@ User: "Actually, here's what I need: [refined instruction with specifics]"
### Context Management
Claude Code operates within a ~200K token context window:
Claude Code operates within a **200K token context window** (1M beta available via API — see [200K vs 1M comparison](line 1751)):
| Component | Approximate Size |
|-----------|------------------|
@ -4292,7 +4329,7 @@ The `.claude/` folder is your project's Claude Code directory for memory, settin
| Personal preferences | `CLAUDE.md` | ❌ Gitignore |
| Personal permissions | `settings.local.json` | ❌ Gitignore |
### 3.27.5 Version Control & Backup
### 3.27.6 Version Control & Backup
**Problem**: Without version control, losing your Claude Code configuration means hours of manual reconfiguration across agents, skills, hooks, and MCP servers.
@ -19597,4 +19634,4 @@ We'll evaluate and add it to this section if it meets quality criteria.
**Contributions**: Issues and PRs welcome.
**Last updated**: January 2026 | **Version**: 3.27.5
**Last updated**: January 2026 | **Version**: 3.27.6