OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Skillcorpus

Skillcorpus is an open-source Python pipeline that collects SKILL.md files from public repositories, applies curation and safety and license gates, evaluates skill quality, and retrieves task-relevant skills as context for AI agents.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Skillcorpus — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Skillcorpus when an agent needs retrieved procedural knowledge at answer time, or when developers need to build and control a curated skill corpus rather than maintain task instructions separately for each host.

Who it is for

  • Developers building a curated skill layer for a domain-specific agent.
  • Teams seeking to add retrieval of task-specific SKILL.md context to an agent host.
  • Maintainers who want to evaluate candidate skills and adapt corpus, retrieval, export or host-plugin behavior.

Documented uses

  • Aggregate skills from public repositories or a chosen source registry.
  • Filter and curate skills using safety and license gates.
  • Evaluate skills and retrieve relevant procedures for an agent task.
  • Export a corpus and connect retrieval to supported agent hosts.
  • Adapt taxonomy, quality and deduplication rules, retrieval, exports, evaluation suites or host plugins.
Before you choose

Requirements & limits

Skillcorpus is open-source. The core code is Apache-2.0 licensed; match and evaluate components are MIT-licensed, and each skill retains its upstream license. The hosted SkillHub can be used without cloning the repository.

  • It is a developer pipeline and retrieval layer, not a standalone agent.
  • Teams can use hosted SkillHub or clone the repository to adapt and operate the open-source machinery.
  • Individual skill licenses remain distinct from the repository's software licenses and are checked as part of curation.
  • Documented integrations include OpenClaw, Hermes, Raven, WorkBuddy and DeepSeek Harness.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Create a curated agent-skill layer

Prepare a retrieval-ready corpus of agent skills from a source registry.

Before you start

  • A SkillCorpus repository clone.
  • A source registry of SKILL.md files.

Steps

  1. Point the pipeline at your source registry.

  2. Apply the curation, safety, and license gates to the collected skills.

  3. Evaluate skill quality and export a corpus for your agents.

Expected outcome

An exported, curated corpus intended to provide task-specific skill context to agents.

Guide limitations

  • Each skill retains its upstream license.
  • The repository describes corpus creation and export, but this recipe does not cover a specific command or a verified run.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
EverMind-AI/SkillCorpus · Official homepage title · retained in its source language
How we review sourcesSuggest a correction