Your agent can do more. Find the skill that makes it happen.
Find your next superpower If you are an agent, refer to our llms.txt for full access."..." for exact phrases · Try typography, debugging or spreadsheetSupports initial repository orientation and broad code localization when direct searches fail. Returns only path and line citations while keeping its file reads outside the main context.
Quick-reference card for caveman modes, skills and commands. Trigger: /caveman-help or "caveman help".
Presents a quick-reference card explaining available Caveman modes, skills, and commands.
Act on a Caveman learn report - review the ranked token sinks, apply cost-lowering fixes with per-edit consent, and report what those fixes returned. Use when asked to lower an agent's token cost, what caveman has saved, to trim a heavy CLAUDE.md, or to offload re-pasted context into cavemem.
Uses Caveman learn reports to identify ranked sources of token consumption and apply cost reductions with consent for each edit. Supports trimming heavy agent instructions, moving repeated context into cavemem, and reporting savings.
Inspect Caveman Cloud's experiment lifecycle and block unsafe execution. Use when asked to start, approve, cancel, promote or roll back a Caveman experiment.
Inspects and manages Caveman Cloud experiments, including starting, approving, canceling, promoting, and rolling back experiments. Includes safeguards against unsafe execution.
Turn a Caveman optimization observation into an operator-chosen candidate with a paired baseline evaluation. Use when asked to inspect or evaluate a Caveman optimization report. Needs explicit approval.
Turns optimization observations into operator-selected candidates paired with baseline evaluations. Supports inspecting and evaluating Caveman optimization reports and requires explicit approval.
Compressed code review - one line per finding with location, problem and fix. Use for /caveman-review, "review this PR", or "review the diff".
Provides compressed reviews of pull requests and diffs. Reports each finding in a single line containing its location, the problem, and a suggested fix.
Wire a repository through the Caveman Cloud gateway so every LLM request is measured, with no behavior change. Use for "set up caveman" or adding LLM spend observability.
A growing collection of real, public skills. Descriptions and instructions indexed 25 Sept 2026.
Skills give your agent reusable instructions for a specific job. Pick one, read what it does, and bring it into your workflow.
A name only tells half the story. Search the full description and instructions to find the right fit.
Read the skill, visit its source, and see exactly what you’re adding to your agent.
Copy the install command from a skill page and run it in your project.
npx skillycli add owner/repo --skill name