agentsclimarketplace

Radin setup hooks

Skill shortcuts/radin/skills/radin-setup-hooks

🐀 the agentic stack for those who have to save tokens

Install
npx -y skills add shortcuts/radin --skill radin-setup-hooks

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 19 days oldThe repository was created 19 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Wire up per-repo hooks/MCP config for radin's companion tools (code-review-graph). Use for "set up hooks", "wire up code-review-graph", "enable the knowledge graph here", or right after install.sh in a new project.

SKILL.md

2.1 KB, 485 tokens by cl100k_base, as published. Nobody here has run it

radin: Set Up Companion-Tool Hooks

install.sh installs companion tool binaries globally (rtk, code-review-graph, caveman). No per-repo config wiring. That config — MCP server registration, hooks, CLAUDE.md instructions — repo-scoped, needs setup once per project. Skill does that setup, for repo user currently in.

Scope

Only code-review-graph needs this step today. caveman = Claude Code plugin — hooks register globally at plugin-install time, nothing repo-scoped to do. rtk = CLI, no hook/MCP wiring. Radin adds more per-repo-wired companion tools later, extend this skill rather than write new one.

Steps

  1. Confirm code-review-graph installed: command -v code-review-graph. Missing: tell user run radin's install.sh first (or install themselves), stop — don't install it from this skill.
  2. Confirm user in repo they want wired (check git rev-parse --show-toplevel, show path). Doesn't look right: ask.
  3. Preview first: run code-review-graph install --platform claude-code --dry-run, show user exactly which files it will write/edit (typically .mcp.json and append to CLAUDE.md/AGENTS.md).
  4. Ask explicit y/n confirmation before writing anything — edits files in user's repo. State plainly: "This will write .mcp.json and append graph-tool instructions to CLAUDE.md in <repo path>. Proceed?"
  5. Yes: run code-review-graph install --platform claude-code -y.
  6. No: stop. Don't run any variant of install command.

Never pass --no-hooks or --no-instructions unless user specifically asks skip one — skill's whole point full wiring.

Non-goals

  • Don't touch global ~/.claude/ state — that's install.sh's job, already done.
  • Don't run against repo user didn't ask about.
  • Don't silently re-run if config already exists. code-review-graph install safe to re-run (idempotent, per own contract), but still confirm with user first, per step 4.

What ships with it: 1 file

137 B alongside SKILL.md

agents/

Gives 0 of the 12 instructions most project setup skills give in 485 tokens

Counted across 999 of the 1,637 authors here whose files we hold, read 2026-08-07

  • Ask one question at a timein 29 of 999, across 28 files
  • Detect the package manager from lockfilesin 28 of 999, across 9 files
  • Present findings to the userin 26 of 999, across 5 files
  • Explore current repo statein 24 of 999, across 3 files
  • Update the agent skills block in place if it existsin 24 of 999, across 3 files
  • Install husky lint-staged and prettierin 23 of 999, across 4 files
  • Create the lintstagedrc filein 22 of 999, across 3 files
  • Commit all changed filesin 22 of 999, across 3 files
  • Run lint-staged to verify it worksin 22 of 999, across 3 files
  • Create the husky pre-commit filein 21 of 999, across 2 files
  • Create a prettierrc file if missingin 21 of 999, across 2 files
  • Initialize huskyin 21 of 999, across 2 files

Said here and by no other author read

  • verify code-review-graph is installed
  • run install with platform and yes flags
  • stop if the user denies confirmation
  • extend this skill for future per-repo tools

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 327,132. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.