eval-dashboard

eval-dashboard is a skill for Claude Code from hoangsonww/AI-News-Briefing. It costs 57 tokens per session (535 once invoked), scanned A, original, MIT.

A single offline web page that turns stored evaluation results into charts and tables. It shows trends, radar charts, distributions, per-card scores, and sortable comparisons; an evaluation measures how well a system performs against test cases.

In plain words
What is it for?
Use it to view evaluation results, compare cards, inspect score changes, and spot failed checks or regressions.
Why use it?
It makes quality results easier to inspect than reading database records or raw files. It can run without a backend or build step.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 eval/export_dashboard.py --judge claude-haiku-4-5-20251001 --open.

Part of the ai-news-briefing plugin — 11 skills, 3 agents, 1 hook, 2 MCP servers shipped together

Good fit Use it to view evaluation results, compare cards, inspect score changes, and spot failed checks or regressions.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/hoangsonww/AI-News-Briefing
agentmods
npx agentmods add skills/hoangsonww/ai-news-briefing/eval-dashboard

Made for: Claude Code.

Or install ai-news-briefing, the plugin that ships this one along with the rest of its 11 skills, 3 agents, 1 hook, 2 MCP servers.

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 eval-dashboard

README.md
[![agentmods](https://agentmods.dev/badge/skills/hoangsonww/ai-news-briefing/eval-dashboard/github.svg)](https://agentmods.dev/skills/hoangsonww/ai-news-briefing/eval-dashboard)
Your own site
<a href="https://agentmods.dev/skills/hoangsonww/ai-news-briefing/eval-dashboard"><img src="https://agentmods.dev/badge/skills/hoangsonww/ai-news-briefing/eval-dashboard/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for eval-dashboard

Your own site · 80×15
<a href="https://agentmods.dev/skills/hoangsonww/ai-news-briefing/eval-dashboard"><img src="https://agentmods.dev/badge/skills/hoangsonww/ai-news-briefing/eval-dashboard.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 535 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00057 $0.00535
Opus 5 $0.00028 $0.00267
Sonnet 5 $0.00011 $0.00107
Haiku 4.5 $0.00006 $0.00053

Measured 12d ago against content hash 023a8ae5068d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

eval-dashboard 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 12d 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.

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-plugins/ai-news-briefing/skills/eval-dashboard/SKILL.md · 45 lines

What it actually says

Eval — Interactive Dashboard

Single-file offline UI over eval/store.sqlite + eval/golden/. No backend, no build step. Chart.js loads from a CDN, but everything else works over file://.

How to invoke

make eval-dashboard                                  # regenerate dashboard/data.js
make eval-dashboard OPEN=1                           # also open in default browser
make eval-dashboard DASHBOARD_JUDGE=claude-haiku-4-5-20251001  # filter rows

Direct invocation:

python3 eval/export_dashboard.py --judge claude-haiku-4-5-20251001 --open

Behavior

  1. Pull rows from eval/store.sqlite (latest-per-date wins).
  2. Join each row with its corresponding eval/golden/<date>.json baseline.
  3. Compute summary stats: composite min/max/median/mean, axis medians, drift z-score, gate-fail count, regression count.
  4. Serialize to eval/dashboard/data.js as window.EVAL_DATA = {...}.
  5. Optionally launch the default browser pointing at eval/dashboard/index.html via file://.

Panels rendered

Panel Visualization
Stat cards Composite median + mean, drift status, gate fails, regressions
Composite trend Line chart with baseline overlay + dashed 3.0 gate threshold
Axis radar 5-axis median across all cards
Composite histogram Buckets < 2.5≥ 4.5
Per-card stacked bars Each card's weighted axis contributions
Per-card table Sortable, filterable (All / Below gate / Regressed / Composite ≥ 4), live search

What to tell the user

After running, tell them the dashboard path (eval/dashboard/index.html) and how many cards / goldens loaded. If OPEN=1 was not passed, give them the open command for their platform (open on macOS, xdg-open on Linux, start on Windows). Mention the dashboard is regenerated from the store — they should re-run this after any make eval-backfill to refresh visualizations.

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. 12d ago First seen · 45 lines · 57 tokens per session scan A 023a8ae5068d

Subscribe to this mod's changes

eval-dashboard is a skill published in the GitHub repository hoangsonww/AI-News-Briefing (42 stars, last pushed 4d ago), licensed MIT. It adds 57 tokens to every session and 535 once invoked, about $0.0003 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 2026-08-30.