Caveman Manage

Caveman-manage is a deliberately conservative skill for Caveman Cloud experiments. It reads experiment state and results, summarizes the evidence (sample sizes, quality, guardrails, cost), and proposes exactly one action, but it won't execute lifecycle changes like start, approve, cancel, or rollback. The source is explicit that current server enforcement isn't atomic yet, so mutations stay blocked and any "approve" recommendation is advisory only. Reach for it when you want a clear, evidence-based read on whether an experiment is promotable, not when you expect it to flip the switch for you. The main thing to know is that absent guardrail data is treated as incomplete, not as a pass, which is the right default even if it means more waiting.

skills.sh installs: 118.7K
GitHub stars: 118.7K
last updated: Oct 26, 2026

View on GitHub

Install

$ npx -y skills add juliusbrussee/caveman --skill caveman-manage --agent claude-code

Installs into .claude/skills of the current project.

Files

Manage eval-gated experiments

Treat every lifecycle change as a production control action. Read current state and results, then report one supported recommendation or block. Current agent MCP is intentionally read-only: control-api does not yet enforce a complete lifecycle transition table and evidence gate atomically.

Non-negotiable gates

A request to review, inspect, explain, or recommend authorizes reads only.
Never approve an experiment whose results are pending, whose required guardrails are absent, or whose evidence reports a breach.
Never convert experiment lift into verified_savings. Only active real traffic plus provider-causal, provider-complete ledger evidence can do that.
Never supply an organization id. Project and tenant scope come from the logged-in Caveman identity and server RBAC.
Never execute a lifecycle mutation, even after user approval. Exact <action>:<experiment_id> strings are agent-generatable and are not proof of human intent.
Unknown states and server errors fail closed. Report exact cave_snake_code.