Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/gxdevs/grouter/installnpx skills add GXDEVS/grouter --skill installgit clone --depth 1 https://github.com/GXDEVS/grouterWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00173 | $0.02593 |
| Opus 5 | $0.00086 | $0.01296 |
| Sonnet 5 | $0.00035 | $0.00519 |
| Haiku 4.5 | $0.00017 | $0.00259 |
Grade F, and why
install scanned grade F with 5 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- Don't run `sudo npm install -g` — grouter is Bun-first, and a sudo install will fight the `bun link` that `setup.sh` uses. Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
| Neither bun nor docker | Stop — install [Bun](https://bun.sh) (`curl -fsSL https://bun.sh/install \| bash`) or Docker first. Tell the user which. | Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
5. **Confirm before actions that touch global state** (global `bun install -g`, `bun link`, `docker compose up`, writing to `~/.claude/settings.json`). Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf ~/.grouter # only if the user explicitly asks Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| Neither bun nor docker | Stop — install [Bun](https://bun.sh) (`curl -fsSL https://bun.sh/install \| bash`) or Docker first. Tell the user which. | How it starts
The opening of the file, as written. The whole thing — 267 lines — stays where its author put it; the contents beside it link to each section on GitHub.
grouter — Install & First-Run Flow
Get a user from zero to a running grouter serve with a working proxy at http://localhost:3099. There are four install paths — pick one with the user, don't run all of them.
Ground rules
- Never
sudo. grouter installs to the user's~/.bun/bin(or runs in a container). If a command seems to need root, stop and ask. - Never overwrite the user's shell config silently. PATH fixes get printed as a one-liner for the user to run, not injected.
- Don't wipe
~/.grouter/. That's where accounts and the SQLite DB live. Uninstall removes the binary only unless the user explicitly asks to drop data. - Verify at each step. After install, confirm
grouter --versionworks before moving to first-run. - Confirm before actions that touch global state (global
bun install -g,bun link,docker compose up, writing to~/.claude/settings.json).
Step 0 — Detect the environment
Run in parallel:
bun --version 2>/dev/null || echo "NO_BUN"
docker --version 2>/dev/null || echo "NO_DOCKER"
command -v grouter && grouter --version 2>/dev/null || echo "NO_GROUTER"
echo "$PATH" | tr ':' '\n' | grep -E "\.bun/bin$" || echo "BUN_BIN_NOT_IN_PATH"
uname -s
Decision:
| State | Recommend |
|---|---|
bun present, user just wants to try |
Path A — bunx |
bun present, user wants permanent install |
Path B — global install |
| Inside a clone of the repo, contributor | Path C — from source |
docker present, no bun, or user prefers containers |
Path D — docker |
| Neither bun nor docker | Stop — install Bun (curl -fsSL https://bun.sh/install | bash) or Docker first. Tell the user which. |
If grouter is already on PATH, ask whether the user wants to upgrade (rerun their original install path) or reinstall fresh (uninstall → install).
Path A — bunx (one-shot, no install)
Best for: "just let me try it".
bunx grouter-auth setup
That's it. No global install, no PATH fiddling. Skip to Step 2 (first-run) — bunx runs the setup wizard directly.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 267 lines · 0 tokens per session scan F 10e35aa4a61e
install is a skill published in the GitHub repository GXDEVS/grouter (55 stars, last pushed 4d ago), licensed MIT. It adds 173 tokens to every session and 2,593 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it F with 5 findings (asks for root, downloads and executes remote code, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.