Skip to content
Back to skills

Khoj

ASecurity

"Use Khoj as a self-hosted AI second-brain server, document

  • 247 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 8, 2026
ai-agentspythongofastapidjangodockerdebugginggitapidatabasefrontend

Works with

  • cli
  • api

Security analysis

A100/100

Pro scans all 20 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill khoj --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Khoj?

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

Security grade badge for Khoj
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-khoj/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-khoj)

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: khoj
description: "Use Khoj as a self-hosted AI second-brain server, document
  indexing/search system, chat/agent app, automation/memory service, or Python
  repository to maintain."
disable-model-invocation: true
metadata:
  disco-role: operating
license: AGPL 3.0
---

# Khoj

Use this repo skill when a task names Khoj or asks for a self-hosted AI second brain that combines document ingestion, semantic search, chat, agents, tools, automations, memory, and a FastAPI/Django Python service.

## Quick Routing

- Use [deployment-api](sub-skills/deployment-api/SKILL.md) for installing Khoj, starting the server, Docker/pip setup, database/auth/admin/model-provider configuration, CLI parser checks, and REST route orientation.
- Use [content-indexing](sub-skills/content-indexing/SKILL.md) for `/api/content`, file upload/sync, parser APIs, Markdown/Org/PDF/DOCX/plaintext/image/GitHub/Notion ingestion, conversion, and stale-index diagnostics.
- Use [search-retrieval](sub-skills/search-retrieval/SKILL.md) for `/api/search`, semantic retrieval, `SearchType`, `SearchResponse`, date/file/word filters, embeddings, reranking, distance thresholds, and no-results debugging.
- Use [chat-agents](sub-skills/chat-agents/SKILL.md) for `/api/chat`, WebSocket/REST chat, conversation history/share/file filters, agents, model providers, slash commands, online/webpage/code/image/diagram/research/operator tools, and voice.
- Use [automations-memory](sub-skills/automations-memory/SKILL.md) for `/api/automation`, cron normalization, scheduler leadership, `/automated_task`, `/api/memories`, memory settings, and user/agent memory scoping.
- Use [development](sub-skills/development/SKILL.md) for modifying this Khoj checkout, choosing focused tests, Django model/migration changes, local dev setup, frontend/docs awareness, and maintainer script boundaries.

## First Checks

- Public package: `khoj`; Python support from package metadata is `>=3.10,<3.13`.
- Server entry point: `khoj`, implemented by `khoj.main:run`.
- Safe parser-only CLI validation: do not assume `khoj --help` is side-effect free; use `sub-skills/deployment-api/scripts/inspect_cli.py` or import `khoj.utils.cli.cli`.
- Safe environment probe: run [check_khoj_environment.py](scripts/check_khoj_environment.py) to check package metadata, imports, CLI parser defaults, and optional Django setup without starting a server.
- Staleness check: read [repo-provenance.md](references/repo-provenance.md) before deciding whether this skill matches a current checkout.

## Common Workflows

1. For a self-hosting or API task, start with `deployment-api`, then jump to the endpoint-specific sub-skill.
2. For “my docs are not searchable,” first use `content-indexing` to validate ingestion, then `search-retrieval` to inspect filters, user isolation, embeddings, and ranking.
3. For “chat or agents are failing,” use `chat-agents`, then cross-link to `deployment-api` for provider/base URL/auth setup or to `search-retrieval` for `/notes` grounding.
4. For reminders, newsletters, repeated research, or persistent memories, use `automations-memory`, then `chat-agents` for generated chat responses.
5. For code changes, use `development` first to select focused tests and identify database/frontend/docs side effects, then use the owning user-facing sub-skill for behavior details.

## Safety Boundaries

- Do not start servers, run migrations against a real database, pull containers, install dependencies, trigger external services, or run expensive chat/eval tests unless the user explicitly asks.
- Do not treat source repo docs, tests, examples, or scripts as runtime dependencies for this skill; use the bundled references and helper scripts here.
- Keep credentials out of examples. Khoj can use OpenAI, Anthropic, Gemini, OpenAI-compatible providers, Notion, GitHub, Twilio, Resend, E2B, and other services, but those require user-provided configuration.
- If a current checkout’s commit, dirty paths, package metadata, or public route/API behavior differs from the provenance snapshot, run a skill refresh before relying on detailed claims.

## Shared References

- [package-overview.md](references/package-overview.md) summarizes Khoj’s major surfaces, package metadata, and cross-skill ownership.
- [troubleshooting.md](references/troubleshooting.md) covers cross-cutting failures that span setup, indexing, search, chat, automations, and development.
- [repo-routing-metadata.json](references/repo-routing-metadata.json) is structured metadata consumed by the managed repo-skills router during import.

Files in this skill

  • SKILL.md4.5 KB
  • references/package-overview.md2.6 KB
  • references/repo-provenance.md1.9 KB
  • references/repo-routing-metadata.json503 B
  • references/troubleshooting.md4.6 KB
  • scripts/check_khoj_environment.py3.2 KB
  • sub-skills/automations-memory/SKILL.md2.3 KB
  • sub-skills/automations-memory/references/automations-api.md5.3 KB
  • sub-skills/automations-memory/references/memory-api.md5.1 KB
  • sub-skills/automations-memory/references/scheduler-runtime.md5.6 KB
  • sub-skills/automations-memory/references/troubleshooting.md7.2 KB
  • sub-skills/automations-memory/scripts/validate_cron.py5.1 KB
  • sub-skills/chat-agents/SKILL.md3.5 KB
  • sub-skills/chat-agents/references/agent-configuration.md9.4 KB
  • sub-skills/chat-agents/references/chat-api.md9.3 KB
  • sub-skills/chat-agents/references/tools-and-commands.md12.3 KB
  • sub-skills/chat-agents/references/troubleshooting.md13.5 KB
  • sub-skills/chat-agents/scripts/inspect_chat_schema.py2.8 KB
  • sub-skills/content-indexing/SKILL.md3.2 KB
  • sub-skills/content-indexing/references/content-api.md7.2 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.