fix: correct multi-agent review findings across guide docs

- architecture.md: fix Asana attribution (not Atlassian), Daniela Amodei
  title (President not CEO), update model refs (3.5→4+), update footer
- cheatsheet.md: update date to Feb 2026, TodoWrite→Tasks API in decision tree
- methodologies.md: update date to Feb 2026
- sync-version.sh: escape dots in sed patterns to prevent URL corruption

Findings from 3-agent parallel code review (Agent Teams experiment).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

guide/architecture.md

@@ -310,7 +310,7 @@ Claude Code offers multiple ways to search code, each with specific strengths:
 - [platform.claude.com/docs](https://platform.claude.com/docs/en/build-with-claude/context-windows) (Tier 1)
 - Observed behavior (Tier 2)
 
-Claude Code operates within a fixed context window (200K tokens for Claude 3.5 Sonnet, varies by model).
+Claude Code operates within a fixed context window (~200K tokens, varies by model).
 
 ### Context Budget Breakdown
 
@@ -949,7 +949,7 @@ app.sendFollowUpMessage('Applied filters: EU, Active');
 
 | Tool | Provider | Capabilities |
 |------|----------|--------------|
-| Asana | Atlassian | Project timelines, task boards |
+| Asana | Asana | Project timelines, task boards |
 | Slack | Salesforce | Message drafting with formatting preview |
 | Figma | Figma | Flowcharts, Gantt charts in FigJam |
 | Amplitude | Amplitude | Analytics charts with interactive filtering |
@@ -1196,7 +1196,7 @@ The core philosophy behind Claude Code:
 
 ### Why It Works
 
-1. **Model capability**: Claude 3.5+ is capable enough to handle routing decisions
+1. **Model capability**: Claude 4+ is capable enough to handle routing decisions
 2. **Reduced latency**: Fewer components = faster response
 3. **Simpler debugging**: When something fails, there's one place to look
 4. **Better generalization**: No hand-coded rules to break on edge cases
@@ -1324,6 +1324,6 @@ Found an error? Have verified new information? Contributions welcome:
 
 ---
 
-**Last updated**: January 2026
-**Claude Code version**: 3.3.x
-**Document version**: 1.0.0
+**Last updated**: February 2026
+**Claude Code version**: v2.1.34
+**Document version**: 1.1.0

guide/cheatsheet.md

@@ -6,7 +6,7 @@
 
 **Written with**: Claude (Anthropic)
 
-**Version**: 3.23.2 | **Last Updated**: January 2026
+**Version**: 3.23.2 | **Last Updated**: February 2026
 
 ---
 
@@ -463,7 +463,7 @@ CLAUDE_CODE_ENABLE_TASKS=false claude
 
 ```
 Simple task       → Just ask Claude
-Complex task      → TodoWrite to plan first
+Complex task      → Tasks API to plan first
 Risky change      → Plan Mode first
 Repeating task    → Create agent or command
 Context full      → /compact or /clear

guide/methodologies.md

@@ -2,7 +2,7 @@
 
 > **Confidence**: Tier 2 — Validated by multiple production reports and official documentation.
 >
-> **Last updated**: January 2026
+> **Last updated**: February 2026
 
 This is a quick reference for 15 structured development methodologies that have emerged for AI-assisted development in 2025-2026. For hands-on practical workflows, see [workflows/](./workflows/).
 

scripts/sync-version.sh

@@ -49,8 +49,10 @@ check_file() {
     else
       # Replace all old 3.x.x versions with current version
       for v in $old_versions; do
+        # Escape dots for sed regex (prevents matching digits in URLs/IDs)
+        local escaped_v=$(echo "$v" | sed 's/\./\\./g')
         # macOS compatible sed
-        sed -i '' "s/$v/$VERSION/g" "$file"
+        sed -i '' "s/$escaped_v/$VERSION/g" "$file"
       done
       echo "✅ $file: updated"
     fi