caveman-manage skill

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

guardrails are absent, or whose evidence reports a breach.

1. A request to review, inspect, explain, or recommend authorizes reads only.
2. Never approve an experiment whose results are pending, whose required

traffic plus provider-causal, provider-complete ledger evidence can do that.

1. Never convert experiment lift into verified_savings. Only active real

logged-in Caveman identity and server RBAC.

1. Never supply an organization id. Project and tenant scope come from the

: strings are agent-generatable and are not proof of human intent.

1. Never execute a lifecycle mutation, even after user approval. Exact

cavesnakecode.

1. Unknown states and server errors fail closed. Report exact

Step 1 — Load project and experiment

Prefer MCP:

caveman_context {}
caveman_experiment_get {"action":"get","experiment_id":"<id>"}
caveman_experiment_get {"action":"results","experiment_id":"<id>"}

Use {"action":"list"} when the user has not named an id.

CLI fallback:

caveman cloud experiments list
caveman cloud experiments show <id>
caveman cloud experiments results <id>

Stop if login, project, experiment, or results are unavailable.

Step 2 — Evaluate evidence

Report:

• current lifecycle state and safety class;
• control and candidate sample sizes;
• quality or eval result;
• latency, error, cost, retry, drop, and escalation guardrails when present;
• evidence cost;
• rollback or hold reason;
• whether result is pending, failed, promotable, or active.

Absence is not a pass. If a required field is absent, state evidence incomplete and do not propose approval.

Step 3 — Propose one action

Allowed actions:

current role may approve;

• start — only from a startable draft or queued state with configured graders;
• approve — only with complete passing evidence and a safety class the

policy path. Current deployments may reject this honestly with cavenotimplemented; never describe that response as a rollback.

• cancel — stop a non-active experiment the user no longer wants;
• rollback — revert an active or harmful change through the server's linked

Show recommendation and id:

Proposed action: approve experiment 7f...
Reason: candidate passed quality and every configured guardrail.
Execution: blocked until server-authoritative lifecycle and evidence gates ship.

Do not treat earlier generic statements such as "manage it" or "do what is best" as mutation approval.

Step 4 — Block unsafe execution

Do not emit or run an executable lifecycle command. Explain that current server does not yet enforce every evidence/state transition atomically. CLI and MCP agent surfaces therefore expose experiment reads only.

Step 5 — Re-read after external operator action

If operator says they executed command, read detail and results again. Report server-observed post-state, audit or result response, and any policy-delivery status returned. Never infer success from operator intent alone.

Use this close:

Action: <action> <experiment-id>
Before: <state>
Server response: <status and cave_snake_code if any>
After: <re-read state>
Basis: experiment evidence only. Verified savings unchanged unless the signed
ledger independently records active, provider-causal real-traffic savings.