Borrowing it
Nothing to install: this file belongs to pymc-labs/decision-hub. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/pymc-labs/decision-hub/main/.claude/commands/test-upload-evals-skill.mdgit clone --depth 1 https://github.com/pymc-labs/decision-hubWrote 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.
[](https://agentmods.dev/commands/pymc-labs/decision-hub/test-upload-evals-skill)<a href="https://agentmods.dev/commands/pymc-labs/decision-hub/test-upload-evals-skill"><img src="https://agentmods.dev/badge/commands/pymc-labs/decision-hub/test-upload-evals-skill/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.
<a href="https://agentmods.dev/commands/pymc-labs/decision-hub/test-upload-evals-skill"><img src="https://agentmods.dev/badge/commands/pymc-labs/decision-hub/test-upload-evals-skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01868 |
| Opus 5 | $0.00000 | $0.00934 |
| Sonnet 5 | $0.00000 | $0.00374 |
| Haiku 4.5 | $0.00000 | $0.00187 |
Grade C, and why
test-upload-evals-skill scanned grade C with 2 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 9d 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.
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.
curl -s "https://pymc-labs--api-dev.modal.run/v1/skills/lfiaschi/bayesian-ab-test/eval-report?semver=<version>" | python3 -m json.tool Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s "https://pymc-labs--api-dev.modal.run/v1/skills/lfiaschi/bayesian-ab-test/eval-report?semver=<version>" | python3 -m json.tool How it starts
The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
End-to-end test of publishing a skill with evals on the dev environment. Run each step sequentially — stop and report if any step fails.
All CLI commands use: DHUB_ENV=dev uv run --package dhub-cli dhub <command>
Prerequisites
-
DB migration: Before publishing, ensure the dev database has all columns:
cd server && DHUB_ENV=dev uv run --package decision-hub-server python -c " from decision_hub.settings import create_settings from decision_hub.infra.database import create_engine, metadata settings = create_settings('dev') engine = create_engine(settings.database_url) metadata.create_all(engine) " -
Deploy server: Deploy the latest server code to dev Modal:
cd server && DHUB_ENV=dev modal deploy modal_app.py
Step 1: Create the test skill
Create all files under <scratchpad>/bayesian-ab-test/.
SKILL.md
---
name: bayesian-ab-test
description: "Bayesian A/B testing skill using PyMC for hypothesis testing on log-normally distributed data"
runtime:
language: python
entrypoint: run_ab_test.py
evals:
agent: claude
judge_model: claude-sonnet-4-5-20250929
---
# Bayesian A/B Test
You are a statistical analysis assistant that performs Bayesian A/B testing.
When asked to run the A/B test, execute the `run_ab_test.py` script and present the results clearly.
## Instructions
1. Run `python run_ab_test.py` to generate the dataset and perform the analysis.
2. Present the output verbatim, ensuring all sections are included:
- Hypotheses (H0 and H1)
- Generated data summary
- Sampling diagnostics (convergence, R-hat, ESS)
- Posterior summary and credible intervals
- Conclusion about the null hypothesis
3. Do NOT modify the script or its output.
pyproject.toml
CRITICAL: requires-python MUST be >=3.11 (not >=3.10) because PyMC 5.27.1 requires Python 3.11+. If set to >=3.10, uv sync will fail in the sandbox with a resolution error. Do NOT include a [build-system] section — it causes uv sync to attempt an editable install which fails.
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.
- 9d ago First seen · 170 lines · 0 tokens per session scan C aabbf6c0f18d
test-upload-evals-skill is a command published in the GitHub repository pymc-labs/decision-hub (99 stars, last pushed 27d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,868 tokens. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
playwright-test-heal
Run all my tests and fix the failing ones.
paul:verify
Guide manual user acceptance testing of recently built features.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
ui-snapshot.template
This prompt was authored for Claude-style slash workflows. In Codex runtime, adapt tool calls as follows.
laravel-playwright
E2E Playwright patterns; use the laravel:e2e-playwright skill exactly as written.
qa
Smoke or browser-walk a running app. Report only. Do not implement. Do not merge.