Crew School — Agent Learning System
Run structured learning sessions that produce real knowledge articles, not plans. Prevents the #1 failure mode: agents that say "I'll research this" and stop.
Quick Start
- Identify gaps — What does your agent need to learn? Check what knowledge files exist vs what's needed.
- Pick a topic — Start with critical gaps that directly impact daily work.
- Spawn a session — Use the Session Template below. The anti-laziness guardrails are load-bearing — don't remove them.
- Verify output — Check line count, source count, and scan for planning language.
Session Template
Use this exact template when spawning learning sessions. Fill in the brackets.
CREW LEARNING SESSION — [ROLE]: [TOPIC]
You are the [ROLE] agent. Your assignment is to research [TOPIC] and produce a comprehensive knowledge article.
### Context
[2-3 sentences: why this topic matters for this agent's role]
### Prerequisites
[List knowledge files to read first, or "None"]
### Research Assignment
Deep research on [TOPIC]. You MUST cover:
1. [Subtopic 1] — [Specific questions]
2. [Subtopic 2] — [Specific questions]
3. [Subtopic 3] — [Specific questions]
### Execution Rules — READ CAREFULLY
- DO NOT PLAN. EXECUTE. If you write "I will..." or "Let me create...", STOP and DO IT instead.
- DO NOT stop after outlining. Every section must contain real research findings.
- Search the web. Perform at least 5 web searches. Read at least 2 full articles.
- Cite sources. Every claim needs a source URL.
- Be specific. Names, numbers, examples > generic advice.
### Output Requirements
Write findings to `knowledge/[filename].md` with this structure:
# [Topic]
> **TL;DR:** [2-3 sentence summary]
> **Applies to:** [Which roles]
> **Prerequisites:** [Other knowledge files]
## Key Takeaways
- [Bullet list of actionable findings]
## [Sections with real findings]
## Practical Application
[Steps, templates, checklists the agent can use immediately]
## Sources
[All URLs cited]
### Minimum Quality Thresholds
- >= 150 lines, >= 1500 words, >= 5 cited sources with URLs
- >= 4 H2 sections with substantive content
- 0 instances of "TODO", "TBD", "I will", "I'll create"
- Must include TL;DR, Key Takeaways, and Practical Application sections
### Completion
Append one line to memory/learning-log.md:
[DATE] | [ROLE] | [TOPIC] | [LINES] | [SOURCES] | [ONE-LINE SUMMARY]
Curriculum Design
For multi-session learning, create a curriculum file. See references/curriculum-design.md for:
- How to audit knowledge gaps
- Sequencing topics by dependency
- Joint sessions for cross-functional learning
- Cadence recommendations (learning vs doing ratio)
Scheduling as a Cron
To automate daily learning, create a curriculum tracker JSON and a cron job that reads it. See references/automation.md for the full pattern including:
- Curriculum JSON schema
- Cron job prompt template
- Progress tracking and auto-advancement
Common Failure Modes & Fixes
| Failure |
Cause |
Fix |
| Agent outputs a plan, not research |
Prompt lacks execution mandate |
Add "DO NOT PLAN. EXECUTE." + minimum line count |
| Shallow output (<100 lines) |
No quality floor |
Add minimum thresholds (150 lines, 5 sources) |
| Generic findings, no data |
No web search requirement |
Add "Perform at least 5 web searches" |
| Session ends after 1 search |
Timeout too short |
Set runTimeoutSeconds >= 300 |
| Agent reads prereqs but doesn't research |
Prereqs consume the whole session |
Limit prereqs to 1-2 files, summarize in prompt |
| Duplicate research across sessions |
No knowledge inventory check |
Audit existing files before assigning topics |
Quality Assessment
After a session, check:
- Line count —
wc -l knowledge/[file].md (target: 150+)
- Source count —
grep -c "http" knowledge/[file].md (target: 5+)
- Planning language —
grep -ci "I will\|I'll create\|TODO\|TBD" knowledge/[file].md (target: 0)
- Structure — Has TL;DR, Key Takeaways, Practical Application sections
- Actionability — Could an agent use this to do better work tomorrow?
1---2name: crew-school3description: Structured learning system for AI agent crews. Design curricula, run research sessions, track progress, and prevent common failure modes (lazy output, planning-without-executing). Use when setting up agent learning, running training sessions, auditing knowledge gaps, or building a curriculum for specialized agents. Works for single agents or multi-agent crews.4---56# Crew School — Agent Learning System78Run structured learning sessions that produce real knowledge articles, not plans. Prevents the #1 failure mode: agents that say "I'll research this" and stop.910## Quick Start11121. **Identify gaps** — What does your agent need to learn? Check what knowledge files exist vs what's needed.132. **Pick a topic** — Start with critical gaps that directly impact daily work.143. **Spawn a session** — Use the Session Template below. The anti-laziness guardrails are load-bearing — don't remove them.154. **Verify output** — Check line count, source count, and scan for planning language.1617## Session Template1819Use this exact template when spawning learning sessions. Fill in the brackets.2021```22CREW LEARNING SESSION — [ROLE]: [TOPIC]2324You are the [ROLE] agent. Your assignment is to research [TOPIC] and produce a comprehensive knowledge article.2526### Context27[2-3 sentences: why this topic matters for this agent's role]2829### Prerequisites30[List knowledge files to read first, or "None"]3132### Research Assignment33Deep research on [TOPIC]. You MUST cover:341. [Subtopic 1] — [Specific questions]352. [Subtopic 2] — [Specific questions]363. [Subtopic 3] — [Specific questions]3738### Execution Rules — READ CAREFULLY39- DO NOT PLAN. EXECUTE. If you write "I will..." or "Let me create...", STOP and DO IT instead.40- DO NOT stop after outlining. Every section must contain real research findings.41- Search the web. Perform at least 5 web searches. Read at least 2 full articles.42- Cite sources. Every claim needs a source URL.43- Be specific. Names, numbers, examples > generic advice.4445### Output Requirements46Write findings to `knowledge/[filename].md` with this structure:4748# [Topic]4950> **TL;DR:** [2-3 sentence summary]51> **Applies to:** [Which roles]52> **Prerequisites:** [Other knowledge files]5354## Key Takeaways55- [Bullet list of actionable findings]5657## [Sections with real findings]5859## Practical Application60[Steps, templates, checklists the agent can use immediately]6162## Sources63[All URLs cited]6465### Minimum Quality Thresholds66- >= 150 lines, >= 1500 words, >= 5 cited sources with URLs67- >= 4 H2 sections with substantive content68- 0 instances of "TODO", "TBD", "I will", "I'll create"69- Must include TL;DR, Key Takeaways, and Practical Application sections7071### Completion72Append one line to memory/learning-log.md:73[DATE] | [ROLE] | [TOPIC] | [LINES] | [SOURCES] | [ONE-LINE SUMMARY]74```7576## Curriculum Design7778For multi-session learning, create a curriculum file. See [references/curriculum-design.md](references/curriculum-design.md) for:79- How to audit knowledge gaps80- Sequencing topics by dependency81- Joint sessions for cross-functional learning82- Cadence recommendations (learning vs doing ratio)8384## Scheduling as a Cron8586To automate daily learning, create a curriculum tracker JSON and a cron job that reads it. See [references/automation.md](references/automation.md) for the full pattern including:87- Curriculum JSON schema88- Cron job prompt template89- Progress tracking and auto-advancement9091## Common Failure Modes & Fixes9293| Failure | Cause | Fix |94|---------|-------|-----|95| Agent outputs a plan, not research | Prompt lacks execution mandate | Add "DO NOT PLAN. EXECUTE." + minimum line count |96| Shallow output (<100 lines) | No quality floor | Add minimum thresholds (150 lines, 5 sources) |97| Generic findings, no data | No web search requirement | Add "Perform at least 5 web searches" |98| Session ends after 1 search | Timeout too short | Set runTimeoutSeconds >= 300 |99| Agent reads prereqs but doesn't research | Prereqs consume the whole session | Limit prereqs to 1-2 files, summarize in prompt |100| Duplicate research across sessions | No knowledge inventory check | Audit existing files before assigning topics |101102## Quality Assessment103104After a session, check:1051. **Line count** — `wc -l knowledge/[file].md` (target: 150+)1062. **Source count** — `grep -c "http" knowledge/[file].md` (target: 5+)1073. **Planning language** — `grep -ci "I will\|I'll create\|TODO\|TBD" knowledge/[file].md` (target: 0)1084. **Structure** — Has TL;DR, Key Takeaways, Practical Application sections1095. **Actionability** — Could an agent use this to do better work tomorrow?
dvcrn