# Rstack Skill Evaluator

> Benchmark agent skills by generating eval cases, comparing skill-guided vs baseline runs, and recording artifacts under skills-test/{skill-name}.

- **Type:** Skill
- **Install:** `agentstack add skill-rstackjs-agent-skills-rstack-skill-evaluator`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [rstackjs](https://agentstack.voostack.com/s/rstackjs)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [rstackjs](https://github.com/rstackjs)
- **Source:** https://github.com/rstackjs/agent-skills/tree/main/dev-skills/rstack-skill-evaluator

## Install

```sh
agentstack add skill-rstackjs-agent-skills-rstack-skill-evaluator
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# rstack-skill-evaluator

A thin repo-specific layer on top of `skill-creator`. For workflow (Test / Improve / Benchmark modes), JSON schemas, grading, and viewer details, defer to `skill-creator`'s own SKILL.md and `references/schemas.md`.

## Targeting a skill

If the user hasn't named a target, ask. Skills live under `skills/` (production) and `dev-skills/` (internal-only).

## Artifact layout

For skill ``, two paths are tracked in git; everything else under `skills-test/` is gitignored:

```plaintext
+--------------------------------------+----------------------------------+
|  Tracked path                        |  Purpose                         |
+--------------------------------------+----------------------------------+
|  skills-test//evals/evals.json |  eval definitions                |
|  skills-test//report.md        |  human-readable run summary      |
+--------------------------------------+----------------------------------+
```

Workspaces, raw run outputs, and fixtures may live anywhere — under `skills-test//` or an OS scratch dir — as long as `report.md` references the path so a reader can find them.

## `report.md`

Committed Markdown summary of the latest run. At minimum cover: setup (model, skill version / commit ref, date), aggregate pass rate / tokens / wall time for `with_skill` vs `without_skill`, per-eval breakdown, and pointers to the raw artifacts.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [rstackjs](https://github.com/rstackjs)
- **Source:** [rstackjs/agent-skills](https://github.com/rstackjs/agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-rstackjs-agent-skills-rstack-skill-evaluator
- Seller: https://agentstack.voostack.com/s/rstackjs
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.