Index of every page on this site: llms.txt
All skills
Run it yourself

Initialize, adopt, and route work for a separate Obsidian knowledge vault through the portable claude-obsidian core. Use for vault setup, scaffolding, workspace selection, cross-project configuration, or choosing the correct wiki sub-skill. Triggers: /wiki, set up wiki, scaffold vault, create knowledge base, adopt this vault, Obsidian vault, second brain setup, persistent wiki.

Use this Skill: https://skilld.dev/gh/agricidaniel/claude-obsidian/wiki

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈97 tokens for metadata: the name and description. ≈1.5k when used: this file. ≈9.7k more on demand in 14 files.

Description uses 4.8% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈387 characters, or 4.8%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Wiki orchestration

Treat the installed product as code and the selected user vault as data. Never use the plugin/product root as a vault, even when the current directory happens to be the product checkout.

Resolve the portable core from this skill's installation and invoke it by absolute path:

CORE=/absolute/product/root/scripts/claude-obsidian.py
python3 "$CORE" --help

Resolve a vault in this order: explicit --vault, CLAUDE_OBSIDIAN_VAULT, the nearest .claude-obsidian.json, then an unambiguous initialized vault at or above the current directory. Fail closed when selection is missing or ambiguous.

Baseline setup requires no network egress. Do not fetch templates, plugins, or remote content unless the user separately approves the destinations and budget.

Set up a vault

Use the deterministic setup commands. Both are dry-run by default.

For a new, separate vault:

python3 "$CORE" init /absolute/path/to/vault \
  --generated-at <ISO-UTC> --operation-id init-reviewed
python3 "$CORE" init /absolute/path/to/vault \
  --generated-at <ISO-UTC> --operation-id init-reviewed \
  --approved-plan-sha256 <reviewed-sha256> --apply

For an existing Obsidian vault:

python3 "$CORE" adopt /absolute/path/to/vault \
  --generated-at <ISO-UTC> --operation-id adopt-reviewed
python3 "$CORE" adopt /absolute/path/to/vault \
  --generated-at <ISO-UTC> --operation-id adopt-reviewed \
  --approved-plan-sha256 <reviewed-sha256> --apply

Before --apply, show the selected path and changed-path preview, then pass the emitted approved_plan_sha256 unchanged. Do not use --force unless the user has reviewed the conflicts and explicitly approved replacement. Setup is non-destructive by default and creates no upstream Git remote.

If the user asks for a domain-specific scaffold, establish the baseline first, then read modes.md. Draft the additional pages and configuration as one operation-level transaction. Never mutate vault files with host Write/Edit tools or an Obsidian transport.

Route operations

Route the user's intent without silently broadening it:

Intent Skill
Ingest supplied sources wiki-ingest
Answer from existing vault knowledge wiki-query
Save a specific conversation result save
Research the public web under a budget autoresearch
Check vault health wiki-lint
Draft a cited glossary or consolidate overlapping notes wiki-ingest, with an explicitly reviewed scope
Roll up log entries wiki-fold
Work with a canvas canvas

Query is read-only. Persistence from a query must be an explicit, separately scoped Save operation. Never capture a transcript or update the hot cache merely because a session ended.

Mutation contract

Read operation-transactions.md before any custom scaffold or mutation. One logical operation must produce one inspected and recoverable claude-obsidian.transaction.v1 bundle. Parallel agents may return drafts and evidence only; the orchestrator merges them and applies once. Every canonical page create or removal includes an active index or MOC update in that bundle; update the overview only when the stable high-level picture changed. Raw source payloads are create-only. There are no automatic commits.

Use provenance.md when initializing or changing source and claim ledgers. Unsupported evidence stays unsupported; never invent a source, quote, date, locator, or confidence.

After a successful apply, report the operation ID and exact changed paths. If the user explicitly wants a Git checkpoint, run it separately:

python3 "$CORE" checkpoint OPERATION_ID --vault /absolute/path/to/vault

On a conflict, re-read and rebuild. On interruption, use transaction recover. Reuse an operation ID only with the identical bundle.

Installation context

Read install-modes.md when installation or host behavior matters. Hooks are optional adapters; portable behavior lives in the core and skills.

Conditional references

Read only the reference needed for the current request:

Think, verify, grow

Before applying, pause once: observe existing state, verify the vault selection and evidence, then choose the smallest reversible operation that satisfies the request. Afterward, report uncertainty and the next useful improvement without performing it automatically.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts28d4 checks · Risk SAFE
  • Gen Agent Trust Hub28d

    The 'wiki' skill provides instructions for an AI agent to manage Obsidian knowledge vaults using a local Python-based CLI tool. It prioritizes security through a mandatory user-review workflow for all vault changes, local-only processing by default, and defensive guidance against prompt injection and unauthorized network access.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

  • ZeroLeaks7mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 2e54556. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Activeupdated last month

README badge

README badge for agricidaniel/claude-obsidian/wiki