Clawdiney Local Setup for Obsidian
Index an Obsidian vault and verify Clawdiney search in Codex or OpenCode, without Ollama.
$19
Works with the AI Agents you already use
Clawdiney Local Setup for Obsidian
Example session with this skill installed
Set up Clawdiney for my Obsidian test vault in Codex CLI using OpenAI embeddings. Show me the exact vault files, commands and API data disclosure, and wait for my approval before indexing. Then connect the MCP server and search for “amber telescope” in Architecture.md.
- Read your context and instructions
- Compiled the clawdiney local setup
Verified example: two-note synthetic vault
This example uses actual results from an OpenAI API and stdio MCP smoke test of Clawdiney 0.2.5 installed from PyPI. Temporary directory paths are shortened for readability. A buyer run must first present the selected vault and commands and wait for approval.
| Check | Observed result |
|---|---|
| Embedding provider | OpenAI text-embedding-3-small, 1536 dimensions |
| Index | 2 of 2 Markdown files processed; 2 chunks indexed |
| Local database | 2 documents, 2 chunks, 3 graph entities, 2 relations |
health_check | “All services healthy”; embeddings OK |
search_brain("amber telescope") | Returned Architecture.md, section “Architecture” |
explore_graph("Architecture.md") | Connected to Deployment.md and tag backend |
The matched source was Architecture.md: amber telescope system. [[Deployment]] #backend. Its linked note Deployment.md was also returned. The MCP health response reported that project-card LLM text and semantic enrichment are unavailable because they require Ollama; reranking was disabled. No Ollama process was available during the test.
For your vault: the agent must report your approved vault path, actual document and chunk counts, MCP health, and an expected note returned through your chosen coding agent. Indexing sends approved Markdown content to your OpenAI account; each search query also uses its embedding API. OpenAI charges are separate.
Connects securely to your tools. The creator never sees your data.
About this skill
Turn a selected Obsidian vault into a searchable source for your coding agent. This skill guides a terminal-capable Codex CLI or OpenCode CLI agent through a checked installation of Clawdiney 0.2.5, OpenAI embeddings, local SQLite indexing, MCP registration, and a real search that returns a note from your vault.
What the paid skill delivers
- Checks Python 3.10+, the vault path, existing MCP settings and database location.
- Shows the exact vault and commands before installation or indexing, then waits for your approval.
- Installs the free MIT Clawdiney package from PyPI in a dedicated virtual environment.
- Configures
text-embedding-3-smallwith 1536 dimensions and a new localbrain.db; it preserves an existing database. - Connects the local MCP server to Codex CLI or OpenCode CLI and verifies
health_checkplussearch_brainagainst a note in the approved vault. - Diagnoses missing keys, invalid vault paths, mismatched model dimensions and MCP startup failures.
Requirements and costs
Bring Python 3.10+, a local Obsidian vault, a terminal-capable Codex CLI or OpenCode CLI, and your own OpenAI API key. The Clawdiney core is free MIT software. OpenAI API usage is billed separately. Indexing sends the selected vault’s Markdown content to OpenAI for embeddings, and search queries are also sent for embedding. Review the vault contents before approving indexing.
Scope
This OpenAI-only setup supports indexing, hybrid search, WikiLink/tag graph navigation and MCP tools without an Ollama process. Project-card LLM text and semantic entity extraction still require Ollama. The purchase covers the guided workflow, not a license to the free core, hosted service, continuous support or future updates. Verified agent flows: Codex CLI and OpenCode CLI.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
The assisted workflow is verified in Codex CLI and OpenCode CLI on macOS. Linux and WSL instructions are provided but have not received equivalent end-to-end tests. It indexes Markdown notes; PDFs, images and canvas drawings are outside this offer. OpenAI API usage and costs require the buyer’s own account. Project-card LLM text and semantic entity extraction require Ollama and are unavailable in this OpenAI-only setup.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 26 days ago
- Accounts
OpenAI API key
- Software
Python 3.10+
Codex CLIorOpenCode CLITerminal accessClawdiney 0.2.5Local Obsidian vault
Clawdiney installs from PyPI.
Permissions
Allowed Hosts
Reads the approved vault to index Markdown; creates a new local SQLite database and updates one agent MCP configuration after approval. Sends selected vault Markdown content and search queries to OpenAI for embeddings. Uses OPENAI_API_KEY only from the buyer’s environment; never stores its value in the skill, command or MCP config.