openevidence-skill

KO | EN

gitlite — search

#react #typescript #llm #android #machine-learning #rust #deep-learning #macos #java #nodejs #awesome #awesome-list #openai #open-source

openevidence-skill

(★ 11)

Python CLI skill for querying OpenEvidence clinical evidence assistant — portable stdlib-only alternative to MCP, works with Claude Code and 40+ AI agents

openevidence-skill Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# openevidence-skill

[![Build & Release Skill](https://github.com/htlin222/openevidence-skill/actions/workflows/release.yml/badge.svg)](https://github.com/htlin222/openevidence-skill/actions/workflows/release.yml)
[![GitHub Release](https://img.shields.io/github/v/release/htlin222/openevidence-skill?include_prereleases&label=skill%20version)](https://github.com/htlin222/openevidence-skill/releases/latest)
[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://raw.githubusercontent.com/htlin222/openevidence-skill/main/LICENSE)
[![Skills Protocol](https://img.shields.io/badge/protocol-vercel--labs%2Fskills-blue)](https://github.com/vercel-labs/skills)
[![Compatible Agents](https://img.shields.io/badge/agents-40%2B-green)](https://github.com/vercel-labs/skills#supported-agents)

> Portable, stdlib-only Python skill that queries OpenEvidence (clinical evidence assistant) without an MCP server. Mirrors the `openevidence-mcp` tool surface 1:1.

## Install

```bash
npx skills add htlin222/openevidence-skill
npx skills add -g htlin222/openevidence-skill              # global
npx skills add htlin222/openevidence-skill --agent claude-code  # specific agent
```

Or grab `openevidence.skill` from [the latest release](https://github.com/htlin222/openevidence-skill/releases/latest) and unzip into your skills directory.

## What it does

One CLI (`scripts/oe.py`) with four subcommands matching the MCP tools:

| MCP tool          | Skill command                              | Purpose                                                  |
| ----------------- | ------------------------------------------ | -------------------------------------------------------- |
| `oe_auth_status`  | `python3 scripts/oe.py auth-status`        | Confirm `cookies.json` still authenticates `/api/auth/me` |
| `oe_history_list` | `python3 scripts/oe.py history`            | Paginated question history with optional search          |
| `oe_article_get`  | `python3 scripts/oe.py article <uuid>`     | Fetch an article, extract answer + figures, save artifacts |
| `oe_ask`          | `python3 scripts/oe.py ask "<question>"`   | Submit a new question, optionally wait, save artifacts   |

Output of every command is a single JSON object on stdout - the same shape the MCP returns to its client. Artifacts (markdown answer, raw payload, citations, BibTeX, figures) are written to `$OE_ARTIFACT_DIR` or `./oe-artifacts/<article_id>/`.

See [`SKILL.md`](SKILL.md) for setup, environment variables, and full command reference.

## Skill structure

```
openevidence-skill/
├── SKILL.md
└── scripts/
    └── oe.py
```

## Setup (one time)

1. Export OpenEvidence cookies from a logged-in browser tab (e.g. Cookie-Editor extension → "Export → JSON"). Save as `cookies.json` next to `SKILL.md`. Cookies are gitignored - never commit them.
2. `python3 scripts/oe.py auth-status` should print `"authenticated": true`.

## Protocol

This skill follows the [vercel-labs/skills](https://github.com/vercel-labs/skills) protocol. Pushing a SemVer tag (e.g. `1.0.0`) triggers a GitHub Action that packages the skill as a `.skill` file and creates a GitHub Release with the artifact attached.

## License

Released under the [Apache License 2.0](LICENSE), matching the upstream [openevidence-mcp](https://github.com/htlin222/openevidence-mcp) project.

openevidence-skill

openevidence-skill Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

openevidence-skill

openevidence-skill Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README