Obsidian Perfect

ReviewAudited by ClawScan on May 10, 2026.

Overview

This looks like a legitimate Obsidian automation skill, but it gives the agent broad read/write/delete control of your vault through an API key that is under-declared in the registry metadata.

Install this only if you intentionally want an AI agent to control your Obsidian vault. Back up the vault first, verify OBSIDIAN_URL points to your local Obsidian REST API, protect the API key, and require explicit confirmation before deletes, moves, bulk edits, or generic Obsidian commands.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If invoked incorrectly, the agent could change or remove notes, folders, tasks, templates, or run broad Obsidian commands in the user's vault.

Why it was flagged

The skill documents tools that can modify, delete, and move vault content, plus a generic Obsidian command execution capability. These are purpose-related but high-impact and not tightly scoped in the provided docs.

Skill content
| `note-update` | Atualizar (append/prepend/replace/section) |
| `note-delete` | Deletar com confirmação obrigatória |
| `folder-delete` | Deletar pasta com confirmação |
| `obsidian-command` | Executar comandos do Obsidian |
Recommendation

Use only with a backed-up vault, require explicit approval for destructive or bulk actions, and avoid giving the agent open-ended command requests.

What this means

A user may not realize from the registry metadata that installation requires granting API-based access to their Obsidian vault.

Why it was flagged

The package declares an Obsidian API key and URL as required, while the registry metadata says there are no required env vars and no primary credential. That under-declares a credential that enables vault access.

Skill content
"required_env": [
  "OBSIDIAN_API_KEY",
  "OBSIDIAN_URL"
]
Recommendation

Treat the Obsidian API key as sensitive, verify the target URL is your local Obsidian instance, and update metadata to declare the required credential and access scope.

What this means

Users have less registry-level information to verify where the runnable tool code came from or what setup it needs.

Why it was flagged

The registry provenance and install metadata are sparse even though the package contains many JavaScript tool files. The provided code shown is purpose-aligned and the static scan is clean, so this is a provenance/install-context note rather than proof of malicious behavior.

Skill content
Source: unknown
Homepage: none

No install spec — this is an instruction-only skill.

Code file presence
142 code file(s)
Recommendation

Install only from a trusted source and compare the package contents with an expected repository or signed release when possible.

What this means

Private journal entries, project notes, or other sensitive vault content could be surfaced to the agent during searches, summaries, or health checks.

Why it was flagged

The skill is designed to read and search vault content, including daily notes. This is expected for Obsidian assistance, but it means private notes may be loaded into the agent context.

Skill content
| `search-text` | Busca full-text no vault |
| `daily-read` | Ler daily de uma data específica |
| `vault-stats` | Estatísticas completas do vault |
Recommendation

Avoid connecting vaults containing sensitive material unless you are comfortable with the agent reading it; consider using a separate vault or limiting the API-accessible content.