alignment-hive

This skill should be read before delegating work to subagents or Workflow agents, unless the user has already named a model. Covers the strengths, cost dynamics, and effort levels of the available models — Claude and GPT families — and the mechanics of routing to GPT models through model-router.

Set up, verify, repair, or uninstall the model-router Claude/GPT routing gateway — binary install, CLIProxyAPI, Codex OAuth, OS service, and Claude Code settings wiring.

This skill should be used when the user asks to "set up the github action", "configure async claude", "enable @claude on issues", "set up claude code action", "github action integration", "autonomous claude on github", or wants Claude to work on issues and PRs autonomously via GitHub Actions.

Generate comprehensive literature review from research proposal. Use this skill when the user asks for a "literature review", "lit review", "find related papers", "search for papers about", "what papers exist on", or wants to understand the research landscape for a topic. Can also be invoked with /lit-review.

This skill should be used when the user asks to "set up remote-kernels", "configure remote kernels", "set up GPU", "configure GPU access", "set up RunPod", "set up vast.ai", "set up Kubernetes GPU pods", "configure cloud GPU", or wants to run code on cloud GPUs for the first time.

Retrieval instructions for searching session history. Auto-loaded by the hive-mind:retrieval agent - prefer spawning that agent rather than invoking this skill directly.

Retrieval instructions for searching session history. Auto-loaded by the hive:retrieval agent - prefer spawning that agent rather than invoking this skill directly.

This skill should be used when the user asks about MATS policies, procedures, or information - including compute access, housing, illness policy, reimbursements, mentor meetings, program schedule, Lighthaven, or any other MATS cohort logistics. Trigger phrases include "MATS handbook", "how do I get compute", "MATS housing", "illness policy", "MATS schedule", "Lighthaven", "where is Lighthaven", or questions about MATS program operations.

This skill should be used when a tool call is denied in autonomous mode. Guides through checking existing permissions for alternative paths, and if none exist, proposing well-scoped permission rules — without it, denied tool calls tend to either block on user input or cause subtasks to be silently skipped.

This skill should be used when the user asks to set up or configure Claude Code permissions, autonomous mode, and/or deno sandbox for a project.

This skill should be used when the user asks to "enable sharing", "disable sharing", "stop sharing", "manage data sharing", "change sharing settings", "data sharing preferences", "consent settings", "opt out", "sharing status", "what am I sharing", or when data sharing state needs attention for the current project during /hive:align.