Voicenotes

Sync and access voice notes from Voicenotes.com. Use when the user wants to retrieve their voice recordings, transcripts, and AI summaries from Voicenotes. Supports fetching notes, syncing to markdown, and searching transcripts.

Security Scan

The skill's behavior (shell scripts that use a VOICENOTES_TOKEN and require jq) is consistent with its stated purpose, but the registry metadata omits required environment and binary declarations — a packaging/information mismatch that should be fixed before trusting the skill.

Purpose & Capability

The name/description (sync voice notes from voicenotes.com) matches the actual code: scripts call api.voicenotes.com to list recordings, fetch user info, and write markdown files. The operations requested (token-based API access, downloading notes, writing markdown) are coherent with the described purpose.

Instruction Scope

SKILL.md and the scripts instruct the agent to call the Voicenotes API, convert JSON to markdown, and write files to an output directory. The scripts only access the VOICENOTES_TOKEN env var, the API endpoint, and the workspace filesystem — all expected for a sync tool. They do not contact unexpected external endpoints or read unrelated system files.

Install Mechanism

There is no install spec (instruction-only plus shipped scripts). The scripts are plain shell with curl/jq usage. No downloads from arbitrary URLs or archive extraction are present.

Credentials

The skill requires a VOICENOTES_TOKEN and the sync script requires jq, but the registry metadata lists no required env vars or binaries. That mismatch means the manifest is incomplete and could mislead users or automated installers about required credentials and tooling. The single token is otherwise proportional for the described API access.

Persistence & Privilege

The skill does not request always:true and does not attempt to modify other skill or system configs. It writes markdown files into the workspace (expected for a sync tool) and does not claim elevated or permanent privileges.

What to consider before installing

This skill appears to do what it says: call voicenotes.com API and write notes as markdown. However, the package metadata omitted required items — the scripts expect you to set VOICENOTES_TOKEN and the sync script requires jq. Before installing: (1) confirm the token is obtained from the legitimate Voicenotes settings URL and only grant the minimum-scoped token possible; (2) ensure jq is installed in the environment; (3) be aware the scripts will write files into the chosen output directory (ensure you trust that location); (4) because the skill source is 'unknown', prefer getting code from the official upstream or verifying the repository/maintainer; and (5) if you rely on automated policy checks, update the skill metadata to declare VOICENOTES_TOKEN and required binaries so installers and reviewers are not misled. These inconsistencies look like sloppy packaging rather than explicit malice, but verify the token's permissions and the upstream source before use.

Voicenotes Integration

Sync voice notes from voicenotes.com into the workspace.

Setup

1. Get access token from: https://voicenotes.com/app?obsidian=true#settings
2. Set environment variable: export VOICENOTES_TOKEN="your-token-here"

Quick Start

# Verify connection
./scripts/get-user.sh | jq .

# Fetch recent notes (JSON)
./scripts/fetch-notes.sh | jq '.data[:3]'

# Sync all notes to markdown files
./scripts/sync-to-markdown.sh --output-dir ./voicenotes

Scripts

fetch-notes.sh

Fetch voice notes as JSON.

./scripts/fetch-notes.sh                    # All notes
./scripts/fetch-notes.sh --limit 10         # Last 10 notes
./scripts/fetch-notes.sh --since 2024-01-01 # Notes since date

get-user.sh

Verify token and get user info.

./scripts/get-user.sh | jq '{name, email}'

sync-to-markdown.sh

Sync notes to markdown files with frontmatter.

./scripts/sync-to-markdown.sh --output-dir ./voicenotes

Output format:

---
voicenotes_id: abc123
created: 2024-01-15T10:30:00Z
tags: [idea, project]
---

# Note Title

## Transcript
The transcribed content...

## Summary
AI-generated summary...

API Reference

Base URL: https://api.voicenotes.com/api/integrations/obsidian-sync

Headers required:

• Authorization: Bearer {token}
• X-API-KEY: {token}

Endpoints:

• GET /user/info - User details
• GET /recordings - List voice notes (paginated)
• GET /recordings/{id}/signed-url - Audio download URL

Data Structure

Each voice note contains:

• recording_id - Unique identifier
• title - Note title
• transcript - Full transcript text
• creations[] - AI summaries, action items, etc.
• tags[] - User tags
• created_at / updated_at - Timestamps
• duration - Recording length in seconds

Tips

• Notes are paginated; check links.next for more pages
• Use --since to fetch only new notes since last sync
• AI creations include summaries, todos, and custom prompts
• Rate limited to ~60 requests/minute