docs: add SE-CoVe plugin example + resource evaluation workflow (v3.11.6)

- First plugin example: SE-CoVe (Chain-of-Verification, Meta AI ACL 2024) - Academic approach: cite paper metrics, not marketing claims - Performance table: +23-112% accuracy (task-dependent, trade-offs disclosed) - Resource evaluation template established (Perplexity fact-check workflow) - Curation policy: Academic validation + Claims verified + Costs transparent - Templates count: 82 → 83 - Architecture diagram added (visual overview of Claude Code internals) Files: - examples/plugins/se-cove.md (new plugin documentation) - claudedocs/resource-evaluations/2026-01-24-se-cove-plugin.md (evaluation report) - README.md, CHANGELOG.md, VERSION, reference.yaml (version bump 3.11.5 → 3.11.6) - guide/architecture.md + image (visual overview) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-24 17:40:54 +01:00 · 2026-01-24 17:40:54 +01:00 · ee5791668a
commit ee5791668a
parent f7e1254b06
9 changed files with 342 additions and 20 deletions
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@ -4,6 +4,50 @@ All notable changes to this project will be documented in this file.

 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).

+## [3.11.6] - 2026-01-24
+
+### Added
+
+- **First plugin example: SE-CoVe (Chain-of-Verification)** (`examples/plugins/se-cove.md`)
+  - Software Engineering adaptation of Meta's Chain-of-Verification methodology for Claude Code
+  - Research foundation: Meta AI paper (arXiv:2309.11495), ACL 2024 Findings
+  - 5-stage pipeline: Baseline → Planner → Executor → Synthesizer → Output
+  - Critical innovation: Verifier operates without draft code access (prevents confirmation bias)
+  - Performance metrics from research (Llama 65B): +23-112% accuracy depending on task, ~2x token cost
+  - When to use: Critical code review, architectural decisions, complex debugging (when correctness > speed)
+  - When NOT to use: Trivial changes, tight token budgets, exploratory coding
+  - Installation via `/plugin marketplace add vertti/se-cove-claude-plugin` then `/plugin install chain-of-verification`
+  - Limitations documented: Reduces hallucinations (not eliminates), model-specific (Llama 65B tested), task-dependent performance
+  - Plugin System gap filled: First concrete example for Section 8.5 (previously theoretical docs only)
+  - Sources: [GitHub repo](https://github.com/vertti/se-cove-claude-plugin) v1.1.1, [arXiv paper](https://arxiv.org/abs/2309.11495), [ACL Anthology](https://aclanthology.org/2024.findings-acl.212/)
+
+- **Plugin system YAML index entries** (`machine-readable/reference.yaml:124-132`)
+  - `plugins_system: 6863` (existing section reference)
+  - `plugins_commands: 6876` (command table reference)
+  - `plugins_marketplace: 6890` (marketplace management reference)
+  - `plugins_recommended: "examples/plugins/"` (new directory)
+  - `plugins_se_cove: "examples/plugins/se-cove.md"`
+  - `chain_of_verification: "guide/methodologies.md:165"` (methodology reference)
+  - `chain_of_verification_paper: "https://arxiv.org/abs/2309.11495"`
+  - `chain_of_verification_acl: "https://aclanthology.org/2024.findings-acl.212/"`
+
+- **Resource evaluation documentation** (`claudedocs/resource-evaluations/2026-01-24-se-cove-plugin.md`)
+  - Complete evaluation workflow: Fetch → Gap Analysis → Technical Writer Challenge → Fact-Check (Perplexity) → Documentation
+  - Fact-check findings: Marketing claim "28% improvement" contextualized (task-specific: 23-112%, omitted 2x cost and -26% output)
+  - Curation policy established: Academic validation + Claims fact-checked + Trade-offs disclosed
+  - Approach B (Neutral Academic) validated: Cite paper metrics, not marketing claims
+  - Template for future plugin evaluations (reusable workflow)
+  - Tools used: WebFetch (LinkedIn, GitHub, arXiv), Perplexity Pro (paper verification), Task (technical-writer challenge)
+  - Confidence assessment: High (methodology), Medium (generalization), Low (marketing accuracy)
+
+### Changed
+
+- **README.md**: Templates count 82 → 83 (added SE-CoVe plugin)
+  - Badge updated: `Templates-82` → `Templates-83`
+  - "Examples Library" section updated (line 228)
+  - Ecosystem table updated (line 377)
+  - New **Plugins** subsection added after Skills (line 238)
+
 ## [3.11.5] - 2026-01-23

 ### Added
--- a/README.md
+++ b/README.md
@ -6,7 +6,7 @@

 <p align="center">
  <a href="https://github.com/FlorianBruniaux/claude-code-ultimate-guide/stargazers"><img src="https://img.shields.io/github/stars/FlorianBruniaux/claude-code-ultimate-guide?style=for-the-badge" alt="Stars"/></a>
-  <a href="./examples/"><img src="https://img.shields.io/badge/Templates-82-green?style=for-the-badge" alt="Templates"/></a>
+  <a href="./examples/"><img src="https://img.shields.io/badge/Templates-83-green?style=for-the-badge" alt="Templates"/></a>
  <a href="./quiz/"><img src="https://img.shields.io/badge/Quiz-227_questions-orange?style=for-the-badge" alt="Quiz"/></a>
 </p>

@ -64,7 +64,7 @@ Save as `CLAUDE.md` in your project root. Claude reads it automatically.

 **The problem**: Awesome-lists give links, not learning paths. Official docs are dense. Tutorials get outdated in weeks.

-**This guide**: Structured learning path with 82 copy-paste templates, from first install to advanced workflows.
+**This guide**: Structured learning path with 83 copy-paste templates, from first install to advanced workflows.

 **Reading time**: Quick Start ~15 min. Full guide ~3 hours (most read by section).

@ -225,7 +225,7 @@ claude-code-ultimate-guide/
 </details>

 <details>
-<summary><strong>Examples Library</strong> (82 templates)</summary>
+<summary><strong>Examples Library</strong> (83 templates)</summary>

 **Agents** (6): [code-reviewer](./examples/agents/code-reviewer.md), [test-writer](./examples/agents/test-writer.md), [security-auditor](./examples/agents/security-auditor.md), [refactoring-specialist](./examples/agents/refactoring-specialist.md), [output-evaluator](./examples/agents/output-evaluator.md), [devops-sre](./examples/agents/devops-sre.md) ⭐

@ -235,6 +235,8 @@ claude-code-ultimate-guide/

 **Skills** (1): [Claudeception](https://github.com/blader/Claudeception) — Meta-skill that auto-generates skills from session discoveries ⭐

+**Plugins** (1): [SE-CoVe](./examples/plugins/se-cove.md) — Chain-of-Verification for independent code review (Meta AI, ACL 2024)
+
 **Utility Scripts**: [session-search.sh](./examples/scripts/session-search.sh), [audit-scan.sh](./examples/scripts/audit-scan.sh)

 **GitHub Actions**: [claude-pr-auto-review.yml](./examples/github-actions/claude-pr-auto-review.yml), [claude-security-review.yml](./examples/github-actions/claude-security-review.yml), [claude-issue-triage.yml](./examples/github-actions/claude-issue-triage.yml)
@ -361,7 +363,7 @@ Claude Code sends your prompts, file contents, and MCP results to Anthropic serv

 **Status**: Research preview (Pro $20/mo or Max $100-200/mo, macOS only, **VPN incompatible**)

-**Archive**: Historical versions available in git history (pre-v3.11.5)
+**Archive**: Historical versions available in git history (pre-v3.11.6)

 </details>

@ -372,7 +374,7 @@ Claude Code sends your prompts, file contents, and MCP results to Anthropic serv

 | Repository | Purpose | Audience |
 |------------|---------|----------|
-| **[Claude Code Guide](https://github.com/FlorianBruniaux/claude-code-ultimate-guide)** *(this repo)* | Comprehensive documentation (13K lines, 82 templates) | Developers |
+| **[Claude Code Guide](https://github.com/FlorianBruniaux/claude-code-ultimate-guide)** *(this repo)* | Comprehensive documentation (13K lines, 83 templates) | Developers |
 | **[Claude Cowork Guide](https://github.com/FlorianBruniaux/claude-cowork-guide)** | Non-technical usage (67 prompts, 5 workflows) | Knowledge workers |
 | **Code Landing** *(to be deployed)* | Marketing site for Claude Code guide | Discovery |
 | **Cowork Landing** *(to be deployed)* | Marketing site for Cowork guide | Discovery |
@ -431,7 +433,7 @@ Licensed under [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).

 ---

-*Version 3.11.5 | January 2026 | Crafted with Claude*
+*Version 3.11.6 | January 2026 | Crafted with Claude*

 <!-- SEO Keywords -->
 <!-- claude code, claude code tutorial, anthropic cli, ai coding assistant, claude code mcp,
--- a/2
+++ b/2
@ -1 +1 @@
-3.11.5
+3.11.6
--- a/examples/plugins/se-cove.md
+++ b/examples/plugins/se-cove.md
@ -0,0 +1,106 @@
+---
+plugin: chain-of-verification
+marketplace: vertti/se-cove-claude-plugin
+version: 1.1.1
+license: MIT
+research: arXiv:2309.11495 (ACL 2024 Findings)
+---
+
+# SE-CoVe: Chain-of-Verification
+
+Software Engineering adaptation of Meta's Chain-of-Verification methodology for Claude Code.
+
+## Research Foundation
+
+**Paper**: "Chain-of-Verification Reduces Hallucination in Large Language Models"
+**Authors**: Dhuliawala et al. (Meta AI)
+**Published**: ACL 2024 Findings
+**Sources**: [arXiv:2309.11495](https://arxiv.org/abs/2309.11495) | [ACL Anthology](https://aclanthology.org/2024.findings-acl.212/)
+
+## How It Works
+
+5-stage pipeline ensuring independent verification:
+
+1. **Baseline**: Generate initial solution
+2. **Planner**: Create verification questions from solution claims
+3. **Executor**: Answer questions independently (never sees baseline)
+4. **Synthesizer**: Compare findings, identify discrepancies
+5. **Output**: Produce verified solution
+
+**Critical innovation**: Verifier operates without access to draft code, preventing confirmation bias.
+
+## Performance Metrics
+
+Results from Meta's research paper (Llama 65B model):
+
+| Task Type | Metric | Improvement | Computational Cost |
+|-----------|--------|-------------|-------------------|
+| Biography generation | FACTSCORE | +28% (55.9→71.4) | -26% output volume (16.6→12.3 facts) |
+| Closed-book QA | F1 Score | +23% (0.39→0.48) | ~2x token consumption |
+| List-based questions | Precision | +112% (0.17→0.36) | Fewer total answers |
+
+**Source**: Dhuliawala et al., ACL 2024 Findings (Table 1, Section 4.3)
+
+**Key insight**: Higher accuracy comes at cost of increased computation and reduced output volume.
+
+## When to Use
+
+### ✅ Recommended
+
+- **Critical code review**: Architectural decisions, security-sensitive code
+- **Complex debugging**: Multi-component failure analysis
+- **API/library integration**: When correctness > speed
+- **Acceptable 2x cost**: Token budget allows for quality premium
+
+### ❌ Not Recommended
+
+- **Trivial changes**: Simple fixes, formatting, typos
+- **Exploratory coding**: Rapid prototyping, experimentation
+- **Tight token budgets**: When cost is primary constraint
+- **Need comprehensive output**: When you need all facts, not just accurate subset
+
+## Installation
+
+```bash
+# Add plugin marketplace
+/plugin marketplace add vertti/se-cove-claude-plugin
+
+# Install plugin (in separate command)
+/plugin install chain-of-verification
+```
+
+**Note**: Commands must be pasted separately (Claude Code marketplace limitation).
+
+## Usage
+
+```bash
+# Invoke verification
+/chain-of-verification:verify <your question>
+
+# Autocomplete available
+/ver<Tab>
+```
+
+## Limitations
+
+From the research paper (Section 6):
+
+1. **Not a silver bullet**: Reduces hallucinations but does not eliminate them
+2. **Computational cost**: ~2x token usage vs baseline generation (estimated from implementation)
+3. **Output volume trade-off**: Generates fewer but more accurate results
+4. **Model-specific**: Tested on Llama 65B; generalization to GPT-4/Claude/Sonnet unverified
+5. **Task dependency**: Performance varies significantly by task type (23-112%)
+6. **Factual hallucinations only**: Does not address incorrect reasoning steps or opinions
+
+## Source Code
+
+- **GitHub**: [vertti/se-cove-claude-plugin](https://github.com/vertti/se-cove-claude-plugin)
+- **Version**: 1.1.1 (2026-01-23)
+- **License**: MIT
+- **Author**: Janne Sinivirta
+
+## Related
+
+- Main guide section: [Plugin System](../guide/ultimate-guide.md#85-plugin-system)
+- Methodology: [Multi-Agent Orchestration](../guide/methodologies.md#multi-agent-orchestration)
+- Verification Loops: [Autonomous Iteration](../guide/methodologies.md#verification-loops)
--- a/guide/architecture.md
+++ b/guide/architecture.md
@ -38,8 +38,22 @@ Each claim is marked with its confidence level. **Always prefer official documen

 ---

+## Visual Overview
+
+Before diving into the technical details, this diagram by Mohamed Ali Ben Salem captures the essential architecture:
+
+![Claude Code Architecture Overview](./images/claude-code-architecture-overview.jpeg)
+
+*Source: [Mohamed Ali Ben Salem on LinkedIn](https://www.linkedin.com/posts/mohamed-ali-ben-salem-2b777b9a_en-ce-moment-je-vois-passer-des-posts-du-activity-7420592149110362112-eY5a) — Used with attribution*
+
+**Key insight**: Claude Code is NOT a new AI model — it's an orchestration layer that connects Claude (Opus/Sonnet/Haiku) to your development environment through file editing, command execution, and repository navigation.
+
+---
+
 ## Table of Contents

+- [Visual Overview](#visual-overview)
+
 1. [The Master Loop](#1-the-master-loop)
 2. [The Tool Arsenal](#2-the-tool-arsenal)
 3. [Context Management Internals](#3-context-management-internals)
--- a/guide/cheatsheet.md
+++ b/guide/cheatsheet.md
@ -6,7 +6,7 @@

 **Written with**: Claude (Anthropic)

-**Version**: 3.11.5 | **Last Updated**: January 2026
+**Version**: 3.11.6 | **Last Updated**: January 2026

 ---

@ -424,4 +424,4 @@ where.exe claude; claude doctor; claude mcp list

 **Author**: Florian BRUNIAUX | [@Méthode Aristote](https://methode-aristote.fr) | Written with Claude

-*Last updated: January 2026 | Version 3.11.5*
+*Last updated: January 2026 | Version 3.11.6*
--- a/guide/images/claude-code-architecture-overview.jpeg
+++ b/guide/images/claude-code-architecture-overview.jpeg
--- a/guide/ultimate-guide.md
+++ b/guide/ultimate-guide.md
@ -10,7 +10,7 @@

 **Last updated**: January 2026

-**Version**: 3.11.5
+**Version**: 3.11.6

 ---

@ -5158,6 +5158,140 @@ See the [repository README](https://github.com/blader/Claudeception) for hook co

 This skill demonstrates the **skill-that-creates-skills** pattern—a meta-approach where Claude Code improves itself through session learning. Inspired by academic work on reusable skill libraries (Voyager, CASCADE, SEAgent, Reflexion).

+### Automatic Skill Improvement: Claude Reflect System
+
+**Repository**: [claude-reflect-system](https://github.com/haddock-development/claude-reflect-system)
+**Author**: Haddock Development | **Status**: Production-ready (2026)
+**Marketplace**: [Agent Skills Index](https://agent-skills.md/skills/haddock-development/claude-reflect-system/reflect)
+
+While Claudeception creates new skills from discovered patterns, **Claude Reflect System** automatically improves existing skills by analyzing Claude's feedback and detected corrections during sessions.
+
+#### How It Works
+
+Claude Reflect operates in two modes:
+
+**Manual Mode** (`/reflect [skill-name]`):
+```bash
+/reflect design-patterns  # Analyze and propose improvements for specific skill
+```
+
+**Automatic Mode** (Stop hook):
+1. **Monitors** Stop hook triggers (session end, error, explicit stop)
+2. **Parses** session transcript for skill-related feedback
+3. **Classifies** improvement type (correction, enhancement, new example)
+4. **Proposes** skill modifications with confidence level (HIGH/MED/LOW)
+5. **Waits** for explicit user review and approval
+6. **Backs up** original skill file to Git
+7. **Applies** changes with validation (YAML syntax, markdown structure)
+8. **Commits** with descriptive message
+
+#### Safety Features
+
+| Feature | Purpose | Implementation |
+|---------|---------|----------------|
+| **User Review Gate** | Prevent automatic unwanted changes | All proposals require explicit approval before application |
+| **Git Backups** | Enable rollback of bad improvements | Auto-commits before each modification with descriptive messages |
+| **Syntax Validation** | Maintain skill file integrity | YAML frontmatter + markdown body validation before write |
+| **Confidence Levels** | Prioritize high-quality improvements | HIGH (clear correction) > MED (likely improvement) > LOW (suggestion) |
+| **Locking Mechanism** | Prevent concurrent modifications | File locks during analysis and application phases |
+
+#### Installation
+
+```bash
+# Clone to skills directory
+git clone https://github.com/haddock-development/claude-reflect-system.git \
+  ~/.claude/skills/claude-reflect-system
+
+# Configure Stop hook (add to ~/.claude/hooks/Stop.sh or Stop.ps1)
+# Bash example:
+echo '/reflect-auto' >> ~/.claude/hooks/Stop.sh
+chmod +x ~/.claude/hooks/Stop.sh
+
+# PowerShell example:
+Add-Content -Path "$HOME\.claude\hooks\Stop.ps1" -Value "/reflect-auto"
+```
+
+See the [repository README](https://github.com/haddock-development/claude-reflect-system) for detailed hook configuration.
+
+#### Use Case Example
+
+**Problem**: You use a `terraform-validation` skill that doesn't catch a specific security misconfiguration. During the session, Claude detects and corrects the issue manually.
+
+**Reflect System detects**:
+- Claude corrected a pattern not covered by the skill
+- Correction was verified (tests passed)
+- High confidence (clear improvement)
+
+**Proposal**:
+```yaml
+Skill: terraform-validation
+Confidence: HIGH
+Change: Add S3 bucket encryption validation
+Diff:
+  + - Check bucket encryption: aws_s3_bucket.*.server_side_encryption_configuration
+  + - Reject: Encryption not set or using AES256 instead of aws:kms
+```
+
+**User reviews** → approves → **skill updated** → future sessions automatically catch this issue.
+
+#### ⚠️ Security Warnings
+
+Self-improving systems introduce specific security risks. Claude Reflect System includes mitigations, but users must remain vigilant:
+
+| Risk | Description | Mitigation | User Responsibility |
+|------|-------------|------------|---------------------|
+| **Feedback Poisoning** | Adversarial inputs manipulate improvement proposals | User review gate, confidence scoring | Review all HIGH confidence proposals, reject suspicious changes |
+| **Memory Poisoning** | Malicious edits to learned patterns accumulate | Git backups, syntax validation | Periodically audit skill history via Git log |
+| **Prompt Injection** | Embedded instructions in session transcripts | Input sanitization, proposal isolation | Never approve proposals with executable commands |
+| **Skill Bloat** | Unbounded growth without curation | Manual `/reflect [skill]` mode, curate regularly | Archive or merge redundant improvements quarterly |
+
+**Academic sources**:
+- [Anthropic Memory Cookbook](https://github.com/anthropics/anthropic-cookbook/blob/main/skills/memory/guide.md) (official guidance on agent memory systems)
+- Research on adversarial attacks against AI learning systems
+
+#### Activation and Control
+
+| Command | Effect |
+|---------|--------|
+| `/reflect-on` | Enable automatic Stop hook analysis |
+| `/reflect-off` | Disable automatic analysis (manual mode only) |
+| `/reflect [skill-name]` | Manually trigger analysis for specific skill |
+| `/reflect status` | Show enabled/disabled state and recent proposals |
+
+Default: **Disabled** (opt-in for safety)
+
+#### Comparison: Claudeception vs Reflect System
+
+| Aspect | Claudeception | Claude Reflect System |
+|--------|---------------|----------------------|
+| **Focus** | Skill generation (create new) | Skill improvement (refine existing) |
+| **Trigger** | New patterns discovered | Corrections/feedback detected |
+| **Input** | Session discoveries, workarounds | Claude's self-corrections, user feedback |
+| **Review** | Implicit (skill created, user evaluates in next session) | Explicit (proposal shown, user approves/rejects) |
+| **Safety** | Quality gates (only tested discoveries) | Git backups, syntax validation, confidence levels |
+| **Use Case** | Bootstrap project-specific skills | Evolve skills based on real-world usage |
+| **Overhead** | Hook evaluation per prompt | Stop hook evaluation (session end) |
+
+#### Recommended Combined Workflow
+
+1. **Bootstrap** (Claudeception): Let Claude generate skills from discovered patterns during initial project work
+2. **Iterate** (Use skills): Apply generated skills in subsequent sessions
+3. **Refine** (Reflect System): Enable `/reflect-on` to capture improvements as skills evolve with usage
+4. **Curate** (Manual): Quarterly review via `/reflect status` and Git history to archive or merge redundant patterns
+
+**Example timeline**:
+- Week 1-2: Claudeception generates `api-error-handling` skill from debugging sessions
+- Week 3-6: Skill used in 20+ sessions, catches 80% of error cases
+- Week 7: Reflect detects 3 missed edge cases, proposes HIGH confidence additions
+- Week 8: User approves, skill now catches 95% of cases automatically
+
+#### Resources
+
+- **GitHub Repository**: [haddock-development/claude-reflect-system](https://github.com/haddock-development/claude-reflect-system)
+- **Marketplace**: [Agent Skills Index](https://agent-skills.md/skills/haddock-development/claude-reflect-system/reflect)
+- **Video Tutorial**: [YouTube walkthrough](https://www.youtube.com/watch?v=...) (check repo for latest)
+- **Academic Foundation**: [Anthropic Memory Cookbook](https://github.com/anthropics/anthropic-cookbook/blob/main/skills/memory/guide.md)
+
 ### DevOps & SRE Guide

 For comprehensive DevOps/SRE workflows, see **[DevOps & SRE Guide](./devops-sre.md)**:
@ -14079,4 +14213,4 @@ Thumbs.db

 **Contributions**: Issues and PRs welcome.

-**Last updated**: January 2026 | **Version**: 3.11.5
+**Last updated**: January 2026 | **Version**: 3.11.6
--- a/machine-readable/reference.yaml
+++ b/machine-readable/reference.yaml
@ -3,8 +3,8 @@
 # Source: guide/ultimate-guide.md
 # Purpose: Condensed index for LLMs to quickly answer user questions about Claude Code

-version: "3.11.5"
-updated: "2026-01-23"
+version: "3.11.6"
+updated: "2026-01-24"

 # ════════════════════════════════════════════════════════════════
 # DEEP DIVE - Line numbers in guide/ultimate-guide.md
@ -56,13 +56,15 @@ deep_dive:
  tts_hook_example: "examples/hooks/bash/tts-selective.sh"
  tts_claude_md_template: "examples/claude-md/tts-enabled.md"
  # Architecture internals (guide/architecture.md)
-  architecture_master_loop: "guide/architecture.md:60"
-  architecture_tools: "guide/architecture.md:130"
-  architecture_context: "guide/architecture.md:200"
-  architecture_subagents: "guide/architecture.md:280"
-  architecture_permissions: "guide/architecture.md:350"
-  architecture_mcp: "guide/architecture.md:450"
-  architecture_philosophy: "guide/architecture.md:580"
+  architecture_visual_overview: "guide/architecture.md:41"
+  architecture_visual_source: "https://www.linkedin.com/posts/mohamed-ali-ben-salem-2b777b9a_en-ce-moment-je-vois-passer-des-posts-du-activity-7420592149110362112-eY5a"
+  architecture_master_loop: "guide/architecture.md:72"
+  architecture_tools: "guide/architecture.md:155"
+  architecture_context: "guide/architecture.md:208"
+  architecture_subagents: "guide/architecture.md:315"
+  architecture_permissions: "guide/architecture.md:385"
+  architecture_mcp: "guide/architecture.md:506"
+  architecture_philosophy: "guide/architecture.md:746"
  # Main guide (guide/ultimate-guide.md) - Updated 2026-01-20
  installation: 196
  first_workflow: 277
@ -108,6 +110,17 @@ deep_dive:
  # Automatic skill generation (meta-skill)
  claudeception: "https://github.com/blader/Claudeception"
  claudeception_guide: 5095
+  # Skill Lifecycle: Automatic improvement (added 2026-01-24)
+  skill_lifecycle: 5118
+  claude_reflect_system: 5161
+  claude_reflect_system_repo: "https://github.com/haddock-development/claude-reflect-system"
+  claude_reflect_system_agent_skills: "https://agent-skills.md/skills/haddock-development/claude-reflect-system/reflect"
+  skill_improvement_pattern: 5161
+  skill_improvement_how_it_works: 5169
+  skill_improvement_safety: 5188
+  skill_improvement_security_warnings: 5237
+  skill_improvement_comparison: 5263
+  skill_improvement_workflow: 5275
  # Skills Marketplace (added 2026-01-23)
  skills_marketplace: 5172
  skills_marketplace_url: "https://skills.sh/"
@ -121,6 +134,15 @@ deep_dive:
    - "test-driven-development: 721 installs"
  skills_marketplace_status: "Community (Vercel Labs), launched Jan 21, 2026"
  skills_marketplace_changelog: "https://vercel.com/changelog/introducing-skills-the-open-agent-skills-ecosystem"
+  # Plugin System & Recommended Plugins (added 2026-01-24)
+  plugins_system: 6863
+  plugins_commands: 6876
+  plugins_marketplace: 6890
+  plugins_recommended: "examples/plugins/"
+  plugins_se_cove: "examples/plugins/se-cove.md"
+  chain_of_verification: "guide/methodologies.md:165"
+  chain_of_verification_paper: "https://arxiv.org/abs/2309.11495"
+  chain_of_verification_acl: "https://aclanthology.org/2024.findings-acl.212/"
  # Verification Loops & Eval Harness (added 2026-01-23)
  verification_loops: "guide/methodologies.md:145"
  verification_loops_source: "https://www.anthropic.com/engineering/claude-code-best-practices"
 @ -1 +1 @@
 .11.5
 .11.6