sipyourdrink-ltd/bernstein

manager

Planning - decompose goals, create tasks via task server.

bernstein-run

Run a verified multi-agent goal with Bernstein. Use when a task is too large for a single agent session: Bernstein decomposes the goal into tasks, spawns CLI coding agents in parallel git worktrees, verifies their output, and merges results. Also use to check…

bernstein-agents

Manage Bernstein agents - list active agents, inspect their output, kill stalled agents, or stream live logs. Use when the user asks about agents, wants to see what an agent is doing, or needs to kill one.

bernstein-alerts

Show active alerts from Bernstein - failed tasks, stalled agents, budget warnings, blocked tasks needing human intervention. Use when the user asks about problems, errors, warnings, or what needs attention.

bernstein-plan

Create and manage multi-step execution plans in Bernstein. Plans decompose complex goals into stages with dependencies. Use when the user wants to plan a complex feature, break down a large task, or review an execution plan before agents start working.

bernstein-quality

Show quality metrics for Bernstein runs - success rates per model, lint/test pass rates, completion time distributions. Use when the user asks about quality, reliability, which model performs best, or pass rates.

bernstein-status

Show Bernstein orchestrator status - active agents, task progress, costs, and alerts. Use when the user asks about orchestrator status, what agents are doing, task progress, how much has been spent, or what's happening with the build.

architect

System design - module boundaries, API contracts, ADRs.

backend

Python server code, APIs, async, strict typing.

ci-fixer

CI failures - read error, minimal fix, verify.

devops

DevOps - Docker, CI/CD, cloud infra, monitoring.

docs

Documentation - README, API docs, ADRs, tutorials.

frontend

React / Next.js UI, state, accessibility.

ml-engineer

ML - training, inference, embeddings, evaluation.

prompt-engineer

LLM prompts - design, evaluate, tune instructions.

qa

Test writing - pytest suites, edge cases, regressions.

resolver

Git merge conflicts - resolve without losing intent.

retrieval

Retrieval - vector DBs, embeddings, hybrid search, reranking.

reviewer

Code review - correctness, tests, merge-readiness.

security

Security review - OWASP, auth, secrets, input validation.

visionary

Ideation - generate bold feature proposals.

vp

Cross-cell coordination - resolve conflicts, decide pivots.

analyst

Evaluate proposals - feasibility, engineering payoff, risk.

bernstein-approve

Review and approve/reject pending tasks or plans in Bernstein. Use when the user asks about approvals, wants to review agent work, or needs to approve/reject a plan before execution begins.

bernstein-cost

Show detailed cost breakdown and budget status for the Bernstein orchestrator. Use when the user asks about spending, budget, cost per model, cost per agent, or wants a cost projection.

bernstein-create-task

Create a new task in the Bernstein orchestrator. Use when the user wants to add a task, delegate work to an agent, file a bug fix, or queue up work for the orchestrator to handle.