regen-eval-baseline

regen-eval-baseline is a skill for Claude Code from sumithr/sumo-qa. It costs 115 tokens per session (1,819 once invoked), scanned A, original, Apache-2.0.

A tool that saves a JSON snapshot of one QA skill's promptfoo evaluation and compares it with an earlier snapshot. Promptfoo is a tool for testing and scoring prompts.

In plain words
What is it for?
Use it to capture a before-and-after evaluation for one skill and store the result in the QA evaluation records.
Why use it?
It makes changes to a skill measurable by showing how its evaluation results differ from the previous baseline.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter. Also seen: reads .claude/ paths; mentions subagents.

Part of the sumo-qa plugin — 27 skills, 2 agents, 3 hooks, 1 MCP server shipped together

Install

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.

agentmods
npx agentmods add skills/sumithr/sumo-qa/regen-eval-baseline
Any agent
npx skills add sumithr/sumo-qa --skill regen-eval-baseline
Clone the repo
git clone --depth 1 https://github.com/sumithr/sumo-qa

Made for: Claude Code.

Or install sumo-qa, the plugin that ships this one along with the rest of its 27 skills, 2 agents, 3 hooks, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for regen-eval-baseline

README.md
[![agentmods](https://agentmods.dev/badge/skills/sumithr/sumo-qa/regen-eval-baseline.svg)](https://agentmods.dev/skills/sumithr/sumo-qa/regen-eval-baseline)
Your own site
<a href="https://agentmods.dev/skills/sumithr/sumo-qa/regen-eval-baseline"><img src="https://agentmods.dev/badge/skills/sumithr/sumo-qa/regen-eval-baseline.svg" alt="Measured on agentmods" height="20"></a>
Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,819 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00115 $0.01819
Opus 5 $0.00057 $0.00910
Sonnet 5 $0.00023 $0.00364
Haiku 4.5 $0.00012 $0.00182

Measured 6d ago against content hash ff4e04aca3a1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

regen-eval-baseline scanned grade A with 0 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/run_baseline.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/regen-eval-baseline/SKILL.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

regen-eval-baseline

Captures a promptfoo run for one sumo-qa skill and stores its JSON output in docs/qa/runs/eval-baselines/ (gitignored). The deterministic work lives in scripts/run_baseline.py; this document is the guide for picking inputs and reading the output.

When to use

Trigger this skill when the user wants a per-skill eval snapshot. Common phrasings: "baseline this skill", "snapshot the eval", "run the eval for skill X", "capture before/after for the rewrite I just made". The user invokes it explicitly with /regen-eval-baseline; it doesn't auto-trigger.

This is single-skill on purpose. Full-sweep regeneration belongs on npm run eval:all, which also reads the same tests/evals/promptfoo/skill-*.yaml files but runs them sequentially without snapshotting.

Inputs

Pass exactly one config selector — either --skill (the base config) or --config (a suffixed scenario / .ab.yaml control / explicit path) — plus an optional --label.

  1. --skill <name> — the base config tests/evals/promptfoo/skill-<name>.yaml. Resolves that file exactly; it never cross-matches a longer suffixed sibling (e.g. --skill reviewing-before-merge drives skill-reviewing-before-merge.yaml, NOT skill-reviewing-before-merge-adversarial.yaml). If the named config is absent the script lists the available configs.
  2. --config <selector> — for a suffixed scenario config or an .ab.yaml A/B control. Accepts a bare stem (skill-reviewing-before-merge-adversarial), a filename including the double suffix (skill-reviewing-before-merge-adversarial.ab.yaml), or a full/relative path. Resolved exactly — a stem composes one filename and that file must exist; the resolver never falls back to a near-named neighbour, so a base name can't accidentally snapshot a suffixed sibling and vice-versa.
  3. --label — short kebab-case tag for the snapshot, defaults to baseline. Past conventions in the dir: baseline, postcut, greenfix, or describe-the-change like removability-gate. The label vocabulary is open-ended and may itself be multi-hyphen, so the snapshot filename separates the slug from the label with a literal __ (double underscore) — <date>-skill-<slug>__<label>.json. Because a validated kebab slug/label can never contain __, the slug/label boundary stays unambiguous even for a multi-hyphen label. The label is what makes two snapshots taken on the same day distinguishable.

Read the full file on GitHub · 78 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 6d ago First seen · 78 lines · 115 tokens per session scan A ff4e04aca3a1

Subscribe to this mod's changes

regen-eval-baseline is a skill published in the GitHub repository sumithr/sumo-qa (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 115 tokens to every session and 1,819 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

debug-optimize-lcp

Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…

ChromeDevTools/chrome-devtools-mcp · 99 tokens

specflow-use

To connect Rosetta with Grid Dynamics SpecFlow MCP; only when SpecFlow is mentioned and the MCP is installed.

griddynamics/rosetta · 27 tokens

planning-and-task-breakdown

Breaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.

shashankswe2020-ux/whoop-mcp · 56 tokens

qa-knowledge

To run QA engineering — requirements/gap analysis, scenario & spec design, test implementation, failure triage — over the QA knowledge base.

griddynamics/rosetta · 32 tokens

ue-mcp-epic-routing

Use when deciding between ue-mcp's native category actions and Epic's wrapped ToolsetRegistry tools (the epic actions, incl. the Blueprint graph DSL) for a task in Unreal. Pulls in when authoring Blueprint graph bodies, or any time both a native action and an epic action could do the job and you need to pick.

db-lyon/ue-mcp · 81 tokens

frontmcp-setup

Use when starting, scaffolding, or organizing a FrontMCP project. Covers creating a new project (CLI scaffold or manual) for Node, Vercel, and other targets; standalone versus Nx-monorepo layout, naming conventions, generators, and dependency rules; composing multiple @App classes, ESM packages, and remote MCP servers…

agentfront/frontmcp · 176 tokens