Skip to main content 首页 创作者 seaworld008 commonly-used-high-value-skills reflect-learn
reflect-learn Self-improvement through conversation analysis. Extracts learnings from corrections and success patterns, proposes updates to agent files or creates new skills. Philosophy: "Correct once, never again." Use when: (1) User explicitly corrects behavior ("never do X", "always Y"), (2) Session ending or context compaction, (3) User requests /reflect, (4) Successful pattern worth preserving.
跳到安装 Skills Marketplace 发现并探索由社区构建的 Agent Skills
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/seaworld008/Commonly-used-high-value-skills --skill reflect-learn命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
下载 Zip 下载中... 同仓库更多 Skills Multi-agent collaboration plugin that spawns N parallel subagents competing on the same task via git worktree isolation. Agents work independently, results are evaluated by metric or LLM judge, and the best branch is merged. Use when: user wants multiple approaches tried in parallel — code optimization, content variation, research exploration, or any task that benefits from parallel competition. Requires: a git repo.
Design production-grade multi-agent orchestration systems. Covers five core patterns (sequential pipeline, parallel fan-out/fan-in, hierarchical delegation, event-driven, consensus), platform-specific implementations, handoff protocols, state management, error recovery, context window budgeting, and cost optimization.
App Store Optimization toolkit for researching keywords, optimizing metadata, and tracking mobile app performance on Apple App Store and Google Play Store.
CLAWDHUB-PUBLISHING-GUIDE.md 8.3 KB seaworld008
seaworld008/Commonly-used-high-value-skills
打开 GitHub 仓库 name reflect-learn description Self-improvement through conversation analysis. Extracts learnings from corrections and success patterns, proposes updates to agent files or creates new skills. Philosophy: "Correct once, never again." Use when: (1) User explicitly corrects behavior ("never do X", "always Y"), (2) Session ending or context compaction, (3) User requests /reflect, (4) Successful pattern worth preserving. zh_description 用于reflect、learn，支持检索、拆解、反思和决策。 version 2.0.0 author Claude Code Toolkit source in-house source_url tags ["learn", "planning", "reflect", "workflow"] created_at 2026-03-04 updated_at 2026-03-20 quality 5 complexity intermediate allowed-tools null
Reflect - Self-Improvement Skill
Quick Reference
Command Action /reflectAnalyze conversation for learnings /reflect onEnable auto-reflection /reflect offDisable auto-reflection /reflect statusShow state and metrics /reflect reviewReview low-confidence learnings /reflect [agent]Focus on specific agent
Core Philosophy
"Correct once, never again."
When users correct behavior, those corrections become permanent improvements encoded into the agent system - across all future sessions.
Workflow
Step 1: Initialize State
Check and initialize state files using the state manager:
python scripts/state_manager.py init
State includes:
reflect-state.yaml - Toggle state, pending reviewsreflect-metrics.yaml - Aggregate metricslearnings.yaml - Log of all applied learnings
Step 2: Scan Conversation for Signals
Use the signal detector to identify learnings:
python scripts/signal_detector.py --input conversation.txt
Signal Confidence Levels
Confidence Triggers Examples HIGH Explicit corrections "never", "always", "wrong", "stop", "the rule is" MEDIUM Approved approaches "perfect", "exactly", accepted output LOW Observations Patterns that worked, not validated
Step 3: Classify & Match to Target Files Map each signal to the appropriate target:
Category Target Files Code Style code-reviewer, backend-developer, frontend-developerArchitecture solution-architect, api-architect, architecture-reviewerProcess CLAUDE.md, orchestrator agentsDomain Domain-specific agents, CLAUDE.md Tools CLAUDE.md, relevant specialistsNew Skill .claude/skills/{name}/SKILL.md
Step 4: Check for Skill-Worthy Signals Some learnings should become new skills rather than agent updates:
Non-obvious debugging (>10 min investigation)
Misleading error (root cause different from message)
Workaround discovered through experimentation
Configuration insight (differs from documented)
Reusable pattern (helps in similar situations)
Quality Gates (must pass all):
Step 5: Generate Proposals Produce output in this format:
# Reflection Analysis
## Session Context
- **Date** : [timestamp]
- **Messages Analyzed** : [count]
- **Focus** : [all agents OR specific agent name]
## Signals Detected
| # | Signal | Confidence | Source Quote | Category |
|---|--------|------------|--------------|----------|
| 1 | [learning] | HIGH | "[exact words]" | Code Style |
| 2 | [learning] | MEDIUM | "[context]" | Architecture |
## Proposed Agent Updates
### Change 1: Update [agent-name]
**Target** : `[file path]`
**Section** : [section name]
**Confidence** : [HIGH/MEDIUM/LOW]
**Rationale** : [why this change]
```diff
--- a/path/to/agent.md
+++ b/path/to/agent.md
@@ -82,6 +82,7 @@
## Section
* Existing rule
+* New rule from learning
Proposed New Skills
Skill 1: [skill-name] Will create : .claude/skills/[skill-name]/SKILL.md
Conflict Check
Commit Message reflect: add learnings from session [date]
Agent updates:
- [learning 1 summary]
New skills:
- [skill-name]: [brief description]
Extracted: [N] signals ([H] high, [M] medium, [L] low confidence)
Review Prompt
Y - Apply all changes and commitN - Discard all changesmodify - Adjust specific changes1,3 - Apply only changes 1 and 3s1 - Apply only skill 1all-skills - Apply all skills, skip agent updates
### Step 6: Handle User Response
**On `Y` (approve):**
1. Apply each change using Edit tool
2. Run `git add` on modified files
3. Commit with generated message
4. Update learnings log
5. Update metrics
**On `N` (reject):**
1. Discard proposed changes
2. Log rejection for analysis
3. Ask if user wants to modify any signals
**On `modify`:**
1. Present each change individually
2. Allow editing the proposed addition
3. Reconfirm before applying
**On selective (e.g., `1,3`):**
1. Apply only specified changes
2. Log partial acceptance
3. Commit only applied changes
### Step 7: Update Metrics
```bash
python scripts/metrics_updater.py --accepted 3 --rejected 1 --confidence high:2,medium:1
Toggle Commands
Enable Auto-Reflection
Disable Auto-Reflection
Check Status
Review Pending
Output Locations Project-level (versioned with repo):
.claude/reflections/YYYY-MM-DD_HH-MM-SS.md - Full reflection.claude/reflections/index.md - Project summary.claude/skills/{name}/SKILL.md - New skills
~/.claude/reflections/by-project/{project}/ - Cross-project~/.claude/reflections/by-agent/{agent}/learnings.md - Per-agent~/.claude/reflections/index.md - Global summary
Memory Integration Some learnings belong in auto-memory (~/.claude/projects/*/memory/MEMORY.md) rather than agent files:
Learning Type Best Target Behavioral correction ("always do X") Agent file Project-specific pattern MEMORY.md Recurring bug/workaround New skill OR MEMORY.md Tool preference CLAUDE.md Domain knowledge MEMORY.md or compound-docs
When a signal is LOW confidence and project-specific, prefer writing to MEMORY.md over modifying agents.
Safety Guardrails
Human-in-the-Loop
NEVER apply changes without explicit user approval
Always show full diff before applying
Allow selective application
Git Versioning
All changes committed with descriptive messages
Easy rollback via git revert
Learning history preserved
Incremental Updates
ONLY add to existing sections
NEVER delete or rewrite existing rules
Preserve original structure
Conflict Detection
Check if proposed rule contradicts existing
Warn user if conflict detected
Suggest resolution strategy
Integration
With /handover If auto-reflection is enabled, PreCompact hook triggers reflection before handover.
With Session Health At 70%+ context (Yellow status), reminders to run /reflect are injected.
Hook Integration (Claude Code) The skill includes hook scripts for automatic integration:
cp hooks/precompact_reflect.py ~/.claude/hooks/
Configure in ~/.claude/settings.json:
{
"hooks" : {
"PreCompact" : [
{
"hooks" : [
{
"type" : "command" ,
"command" : "uv run ~/.claude/hooks/precompact_reflect.py --auto"
}
]
}
]
}
}
Portability This skill works with any LLM tool that supports:
File read/write operations
Text pattern matching
Git operations (optional, for commits)
Configurable State Location
export REFLECT_STATE_DIR=/path/to/state
No Task Tool Dependency Unlike the previous agent-based approach, this skill executes directly without spawning subagents. The LLM reads SKILL.md and follows the workflow.
Git Operations Optional Commits are wrapped with availability checks - if not in a git repo, changes are still saved but not committed.
Troubleshooting
Session may not have had corrections
Try /reflect review to check pending items
Review the existing rule cited
Decide if new rule should override
Can modify before applying
Check agent name spelling
Use /reflect status to see available targets
May need to create agent file first
File Structure reflect/
├── SKILL.md # This file
├── scripts/
│ ├── state_manager.py # State file CRUD
│ ├── signal_detector.py # Pattern matching
│ ├── metrics_updater.py # Metrics aggregation
│ └── output_generator.py # Reflection file & index generation
├── hooks/
│ ├── precompact_reflect.py # PreCompact hook integration
│ ├── settings-snippet.json # Settings.json examples
│ └── README.md # Hook configuration guide
├── references/
│ ├── signal_patterns.md # Detection rules
│ ├── agent_mappings.md # Target mappings
│ └── skill_template.md # Skill generation
└── assets/
├── reflection_template.md # Output template
└── learnings_schema.yaml # Schema definition
