[2.19.0] Add learnings discovery - spawn sub-agent per learning
- Discover documented learnings from /workflows:compound - Project docs/solutions/ - User ~/.claude/docs/ - Plugin docs/ - Spawn sub-agent for EACH learning to check if relevant to plan - All learning sub-agents run in parallel - Categories: performance-issues, debugging-patterns, config-fixes, etc. - Added learnings to synthesis sources - Fixed section numbering (now 1-9) Institutional knowledge - prevents repeating past mistakes. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
@@ -142,7 +142,62 @@ Task general-purpose: "Use the security-patterns skill at ~/.claude/skills/secur
|
||||
|
||||
**No limit on skill sub-agents. Spawn one for every skill that could possibly be relevant.**
|
||||
|
||||
### 3. Launch Per-Section Research Agents
|
||||
### 3. Discover and Apply Learnings/Solutions
|
||||
|
||||
<thinking>
|
||||
Check for documented learnings from /workflows:compound. These are solved problems that may apply to the plan. Spawn a sub-agent for each learning to check if it's relevant.
|
||||
</thinking>
|
||||
|
||||
**Step 1: Discover all documented learnings**
|
||||
|
||||
```bash
|
||||
# Project-level learnings (from compound-docs skill)
|
||||
find docs/solutions -name "*.md" 2>/dev/null
|
||||
find .claude/docs -name "*.md" 2>/dev/null
|
||||
|
||||
# User's global learnings
|
||||
find ~/.claude/docs -name "*.md" 2>/dev/null
|
||||
|
||||
# Any learnings in the compound-engineering plugin
|
||||
find ~/.claude/plugins/cache/*/compound-engineering/*/docs -name "*.md" 2>/dev/null
|
||||
```
|
||||
|
||||
**Step 2: For each learning found, spawn a sub-agent to check relevance**
|
||||
|
||||
For EACH learning markdown file:
|
||||
```
|
||||
Task general-purpose: "Read this documented learning/solution:
|
||||
|
||||
[Read the learning file content]
|
||||
|
||||
Check if this learning applies to ANY part of this plan:
|
||||
|
||||
[plan content]
|
||||
|
||||
If the learning is relevant:
|
||||
- Explain how it applies
|
||||
- Extract the key insight or solution
|
||||
- Suggest how to incorporate it into the plan
|
||||
|
||||
If not relevant, just say 'Not applicable' and why."
|
||||
```
|
||||
|
||||
**Spawn ALL learning sub-agents in PARALLEL:**
|
||||
- 1 sub-agent per learning file
|
||||
- Each checks if its learning applies to the plan
|
||||
- All run simultaneously
|
||||
|
||||
**Categories of learnings to check:**
|
||||
- `performance-issues/` - Performance optimizations that worked
|
||||
- `debugging-patterns/` - Debugging approaches that solved problems
|
||||
- `configuration-fixes/` - Config issues and their solutions
|
||||
- `integration-issues/` - Third-party integration lessons
|
||||
- `deployment-issues/` - Deployment and production learnings
|
||||
- Any other category directories found
|
||||
|
||||
**These learnings are institutional knowledge - previous solved problems that may prevent repeating mistakes.**
|
||||
|
||||
### 4. Launch Per-Section Research Agents
|
||||
|
||||
<thinking>
|
||||
For each major section in the plan, spawn dedicated sub-agents to research improvements. Use the Explore agent type for open-ended research.
|
||||
@@ -172,7 +227,7 @@ mcp__plugin_compound-engineering_context7__query-docs: Query documentation for s
|
||||
|
||||
Search for recent (2024-2025) articles, blog posts, and documentation on topics in the plan.
|
||||
|
||||
### 4. Discover and Run ALL Review Agents
|
||||
### 5. Discover and Run ALL Review Agents
|
||||
|
||||
<thinking>
|
||||
Dynamically discover every available agent and run them ALL against the plan. Don't filter, don't skip, don't assume relevance. 40+ parallel agents is fine. Use everything available.
|
||||
@@ -238,7 +293,7 @@ Task [agent-name]: "Review this plan using your expertise. Apply all your checks
|
||||
|
||||
Research agents (like `best-practices-researcher`, `framework-docs-researcher`, `git-history-analyzer`, `repo-research-analyst`) should also be run for relevant plan sections.
|
||||
|
||||
### 5. Wait for ALL Agents and Synthesize Everything
|
||||
### 6. Wait for ALL Agents and Synthesize Everything
|
||||
|
||||
<thinking>
|
||||
Wait for ALL parallel agents to complete - skills, research agents, review agents, everything. Then synthesize all findings into a comprehensive enhancement.
|
||||
@@ -247,10 +302,11 @@ Wait for ALL parallel agents to complete - skills, research agents, review agent
|
||||
**Collect outputs from ALL sources:**
|
||||
|
||||
1. **Skill-based sub-agents** - Each skill's full output (code examples, patterns, recommendations)
|
||||
2. **Research agents** - Best practices, documentation, real-world examples
|
||||
3. **Review agents** - All feedback from every reviewer (architecture, security, performance, simplicity, etc.)
|
||||
4. **Context7 queries** - Framework documentation and patterns
|
||||
5. **Web searches** - Current best practices and articles
|
||||
2. **Learnings/Solutions sub-agents** - Relevant documented learnings from /workflows:compound
|
||||
3. **Research agents** - Best practices, documentation, real-world examples
|
||||
4. **Review agents** - All feedback from every reviewer (architecture, security, performance, simplicity, etc.)
|
||||
5. **Context7 queries** - Framework documentation and patterns
|
||||
6. **Web searches** - Current best practices and articles
|
||||
|
||||
**For each agent's findings, extract:**
|
||||
- [ ] Concrete recommendations (actionable items)
|
||||
@@ -261,6 +317,7 @@ Wait for ALL parallel agents to complete - skills, research agents, review agent
|
||||
- [ ] Edge cases discovered (handling strategies)
|
||||
- [ ] Documentation links (references)
|
||||
- [ ] Skill-specific patterns (from matched skills)
|
||||
- [ ] Relevant learnings (past solutions that apply - prevent repeating mistakes)
|
||||
|
||||
**Deduplicate and prioritize:**
|
||||
- Merge similar recommendations from multiple agents
|
||||
@@ -268,7 +325,7 @@ Wait for ALL parallel agents to complete - skills, research agents, review agent
|
||||
- Flag conflicting advice for human review
|
||||
- Group by plan section
|
||||
|
||||
### 6. Enhance Plan Sections
|
||||
### 7. Enhance Plan Sections
|
||||
|
||||
<thinking>
|
||||
Merge research findings back into the plan, adding depth without changing the original structure.
|
||||
@@ -305,7 +362,7 @@ Merge research findings back into the plan, adding depth without changing the or
|
||||
- [Documentation URL 2]
|
||||
```
|
||||
|
||||
### 5. Add Enhancement Summary
|
||||
### 8. Add Enhancement Summary
|
||||
|
||||
At the top of the plan, add a summary section:
|
||||
|
||||
@@ -326,7 +383,7 @@ At the top of the plan, add a summary section:
|
||||
- [Important finding 2]
|
||||
```
|
||||
|
||||
### 6. Update Plan File
|
||||
### 9. Update Plan File
|
||||
|
||||
**Write the enhanced plan:**
|
||||
- Preserve original filename
|
||||
|
||||
Reference in New Issue
Block a user