Bullpen Learning — Productivity skill for Claude Code
Switch self-learning on or off.
How to install Bullpen Learning
Installs to ~/.claude/skills/manavarya09-bullpen-bullpen-learning/SKILL.md
mkdir -p ~/.claude/skills/manavarya09-bullpen-bullpen-learning && curl -fsSL https://raw.githubusercontent.com/Manavarya09/bullpen/HEAD/commands/bullpen-learning.md -o ~/.claude/skills/manavarya09-bullpen-bullpen-learning/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Bullpen Learning does
description: Switch self-learning on or off. When on, the matching Intern logs durable learnings (decisions, patterns, preferences) to memory after each task. Off by default. argument-hint: on | off | status
You are the **bullpen learning toggle**. Read `$ARGUMENTS`:
on→ setlearning: "on"in~/.bullpen/config.json. Confirm: "✓ Learning on. Pip and the other interns will start writing down what your team learns."off→ setlearning: "off". Confirm: *"✓ Learning off. T
Alternatives in Productivity
- C-level Advisor — Executive advisory skills for C-level decisions 5.3k ★
- Schedule Logs — View execution logs for scheduled tasks 510 ★
- Browser Harness JS — Self-healing browser harness that enables LLMs to complete any task 480 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Workflow Maintenance
Recurring workflow maintenance — capture live self-improvements, curate learnings, check trunk health. Idempot
Self Regulated Learning
Self-regulated learning strategies
Ccc Compound
Post-task learning capture — extract patterns, corrections, and decisions to compound productivity
Bullpen
Dispatch a task to your bullpen team. Sam (the orchestrator) picks one specialist or composes a parallel team
Ash
Hand a task directly to Pebble, a Backend Intern.
Task Complete
Validate task completion with multi-stage verification (22 verify- agents), run all checks, and archive with l
Related Agents
Learning Scribe
Gate-7 capture agent. Converts every deep-review fix or observable incident into a durable docs/learnings/ ent
Runtime Observer.Agent
Use when Architrave needs runtime/product truth: health, logs, versions, deployed digests, app launch, or drif
Backend Intern
Use this agent when the user types /ash or asks for backend intern work — e.g., log learnings from the BE work