Skip to content
Back to skills

Integrations And Extensions

ASecurity

"Handle code-review-graph custom languages, embeddings, wiki,

  • 247 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 8, 2026
toolspythongocode-reviewgitsecurity

Security analysis

A100/100

Pro scans all 7 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill integrations-and-extensions --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Integrations And Extensions?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Integrations And Extensions
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-integrations-and-extensions/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-integrations-and-extensions)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: integrations-and-extensions
description: "Handle code-review-graph custom languages, embeddings, wiki,
  registry, daemon, GitHub Action, and eval workflows."
disable-model-invocation: true
metadata:
  disco-role: operating
license: MIT
---

# Integrations and Extensions

Use this sub-skill for advanced or optional workflows: custom parser languages, semantic embeddings, wiki generation, multi-repo registry and daemon support, GitHub Action integration, and evaluation/benchmark guidance.

## Start here

1. Confirm the requested feature is truly optional or integration-focused.
2. Read the narrow reference for the exact surface.
3. Install only the extra needed for that surface, not `all` by default.

Read [references/custom-languages.md](references/custom-languages.md) for `.code-review-graph/languages.toml` workflows. Read [references/embeddings.md](references/embeddings.md) for semantic-search and provider behavior. Read [references/registry-and-daemon.md](references/registry-and-daemon.md) for multi-repo registry, watch daemon, and repo listing. Read [references/github-action.md](references/github-action.md) for PR review automation. Read [references/eval.md](references/eval.md) for benchmark/reproduction guidance. Read [references/troubleshooting.md](references/troubleshooting.md) when an optional dependency, provider, grammar, daemon state, or eval run fails.

## Route by task

| User task | Do this |
| --- | --- |
| "Add a custom language" | Edit `.code-review-graph/languages.toml` and rebuild the graph. |
| "Enable semantic search" | Install the embeddings extra, then run the embedding workflow. |
| "Generate wiki pages" | Use the wiki generation workflow after communities exist. |
| "Register multiple repos" | Use the registry/daemon commands. |
| "Run a PR review in CI" | Use the GitHub Action pattern and the bundled PR renderer. |
| "Reproduce benchmarks" | Follow the eval reference and be prepared for expensive, repo-clone-heavy work. |

## Optional dependency policy

Only install an optional extra when the task requires it:

- `code-review-graph[embeddings]` for local sentence-transformers embeddings.
- `code-review-graph[google-embeddings]` for Google Gemini embeddings.
- `code-review-graph[communities]` for igraph-based community detection.
- `code-review-graph[wiki]` for Ollama-backed wiki summaries.
- `code-review-graph[eval]` for benchmark and report helpers.
- `code-review-graph[enrichment]` for deeper Python call-resolution enrichment.

Do not install `all` unless the user explicitly wants every optional feature in one environment.

## Safe starting points

- Custom language users should start with a small `.code-review-graph/languages.toml` and a rebuild.
- Embedding users should confirm provider/environment variables before enabling cloud providers.
- Registry/daemon users should ensure each registered repository has a valid graph or watchable checkout.
- GitHub Action users should prefer the split analysis/comment workflow for fork safety.

## Boundaries

- For installation and server setup, use `install-and-setup`.
- For diff review and PR comments, use `review-changes`.
- For structural exploration, search, and refactor previews, use `graph-exploration`.

## Verification anchors

Native tests that ground this route include custom-language tests, embeddings tests, registry and daemon tests, wiki tests, eval tests, and PR workflow security tests. Some of these are optional or service-backed; record unverified optional capabilities explicitly instead of implying full support.

Files in this skill

  • SKILL.md3.5 KB
  • references/custom-languages.md1.9 KB
  • references/embeddings.md2 KB
  • references/eval.md1.1 KB
  • references/github-action.md1.5 KB
  • references/registry-and-daemon.md1.7 KB
  • references/troubleshooting.md3.3 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.