Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Edit Markdown Notes Via Ssh

BSecurity

Read and edit Markdown notes on your personal computer via SSH tunnel. Use when the user asks to read, create, or append to notes in their vault.

19 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsgoshellbashgitsecurity

Works with

cli

Security Analysis

B75/100
criticalAccesses sensitive system or user directories

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add rondoflow/rondoflow --skill edit-markdown-notes-via-ssh --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Edit Markdown Notes Via Ssh?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Edit Markdown Notes Via Ssh
[![Security: B — Skills Directory](https://www.skillsdirectory.com/api/skills/rondoflow-edit-markdown-notes-via-ssh/badge)](https://www.skillsdirectory.com/skills/rondoflow-edit-markdown-notes-via-ssh)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: edit-markdown-notes-via-ssh
description: "Read and edit Markdown notes on your personal computer via SSH tunnel. Use when the user asks to read, create, or append to notes in their vault."
category: "Productivity"
author: community
version: "1.2.6"
icon: check-square
---

# Headless Vault CLI

Access Markdown notes on your personal computer from this VPS-hosted bot via SSH tunnel.

**Terminology**: "Local machine" = your personal computer (macOS or Linux) where your notes live. This skill runs on the VPS and connects to your machine via a reverse SSH tunnel.

## Prerequisites

This is an instruction-only skill. Before using it, the user must complete a one-time setup on their local machine:

1. **Install vaultctl** on the local machine (see [setup instructions](https://github.com/logancyang/headless-vault-cli))
2. **Configure SSH forced-command** on the local machine's `~/.ssh/authorized_keys` to restrict the VPS key to only run `vaultctl` (see Security Model below)
3. **Start a reverse SSH tunnel** from the local machine to the VPS, exposing `localhost:2222`
4. **Set the environment variable** `VAULT_SSH_USER` to the local machine's username

## Security Model

This skill connects to the local machine over a pre-configured reverse SSH tunnel. Access is restricted by design:

- **Forced-command restriction**: The VPS SSH key is added to the local machine's `~/.ssh/authorized_keys` with a forced-command wrapper, so the VPS can ONLY execute `vaultctl` — no interactive shell, no arbitrary commands (`rm`, `curl`, etc.)
- **Vault sandboxing**: `vaultctl` validates all file paths are inside `VAULT_ROOT` and rejects path traversal attempts (`..`, symlinks outside vault)
- **Non-destructive**: Only `create` (new files) and `append` (existing files) are supported — no delete, rename, move, or overwrite
- **No credentials stored**: SSH authentication uses the VPS's existing SSH keypair; no additional secrets are stored by this skill

Example `authorized_keys` entry on the local machine:
```
command="/usr/local/bin/vaultctl-wrapper",no-port-forwarding,no-X11-forwarding,no-agent-forwarding ssh-ed25519 AAAA... vps-key
```

This ensures the VPS can only run `vaultctl` commands, even if the tunnel is compromised.

## Available Commands

You have access to these commands ONLY. Do not attempt commands not listed here (no rename, delete, move, or edit commands exist).

| Command | Description |
|---------|-------------|
| `tree` | List vault directory structure |
| `resolve` | Find note by path or title |
| `info` | Get file metadata (lines, bytes, sha256, mtime) |
| `read` | Read note content |
| `create` | Create a NEW note (fails if file exists) |
| `append` | Append content to EXISTING note |

## How to Run Commands

All commands are executed via SSH:
```bash
ssh -4 -p ${VAULT_SSH_PORT:-2222} ${VAULT_SSH_USER}@${VAULT_SSH_HOST:-localhost} vaultctl <command> [args]
```

Always use `-4` to force IPv4 (avoids IPv6 timeout issues).

## Environment Variables

These must be set in the skill's runtime environment on the VPS:

| Variable | Required | Default | Description |
|----------|----------|---------|-------------|
| `VAULT_SSH_USER` | Yes | — | Local machine username for SSH tunnel |
| `VAULT_SSH_PORT` | No | `2222` | SSH tunnel port on localhost |
| `VAULT_SSH_HOST` | No | `localhost` | SSH tunnel host |

## Command Reference

### tree - List vault structure
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree --depth 2
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree --all
```
Options:
- `--depth N` - Maximum depth to traverse
- `--all` - Include all files, not just .md

### resolve - Find note by path or title

**ALWAYS use `--base64`** for path and title arguments to prevent shell injection:
```bash
# echo -n "Projects/Plan.md" | base64 → UHJvamVjdHMvUGxhbi5tZA==
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl resolve --path UHJvamVjdHMvUGxhbi5tZA== --base64

# echo -n "Meeting Notes" | base64 → TWVldGluZyBOb3Rlcw==
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl resolve --title TWVldGluZyBOb3Rlcw== --base64
```

### info - Get file metadata

**ALWAYS use `--base64`** for the path argument:
```bash
# echo -n "Projects/Plan.md" | base64 → UHJvamVjdHMvUGxhbi5tZA==
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl info UHJvamVjdHMvUGxhbi5tZA== --base64
```
Returns JSON: `{"path": "...", "lines": N, "bytes": N, "sha256": "...", "mtime": N}`

### read - Read note content

**ALWAYS use `--base64`** for the path argument:
```bash
# echo -n "Projects/Plan.md" | base64 → UHJvamVjdHMvUGxhbi5tZA==
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl read UHJvamVjdHMvUGxhbi5tZA== --base64
```
Returns JSON: `{"path": "...", "content": "..."}`

### create - Create a NEW note
**IMPORTANT**: Use `--base64` flag with BOTH path AND content base64 encoded. This is required for paths/content with spaces or special characters.

```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl create <base64_path> <base64_content> --base64
```

Example to create "Notes/Morning Brief.md" with content "# Hello\n\nWorld":
```bash
# Encode path: echo -n "Notes/Morning Brief.md" | base64 → Tm90ZXMvTW9ybmluZyBCcmllZi5tZA==
# Encode content: echo -n "# Hello\n\nWorld" | base64 → IyBIZWxsbwoKV29ybGQ=
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl create Tm90ZXMvTW9ybmluZyBCcmllZi5tZA== IyBIZWxsbwoKV29ybGQ= --base64
```

- Creates parent directories automatically
- Fails if file already exists (use `append` to add to existing files)
- File must have `.md` extension
- **NEVER duplicate the title as a heading inside the note content** (e.g., for "My Note.md", don't start content with "# My Note")

### append - Append to EXISTING note
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl append <base64_path> <base64_content> --base64
```

- Fails if file does not exist (use `create` for new files)

## What You CANNOT Do

These operations are NOT supported:
- **Rename** files or folders
- **Delete** files or folders
- **Move** files between folders
- **Edit** specific parts of a file (only append to end)
- **Create** folders without a file (folders are created automatically with `create`)

## Tips

- Always run `vaultctl tree` first to see what notes exist
- Use `vaultctl resolve --title <base64> --base64` to find a note by name
- All output is JSON
- The local machine must be online with tunnel running
- **ALWAYS use `--base64`** for ALL path and content arguments — this is mandatory for security, not optional

## Examples

**Important**: Always run `tree` first if you're unsure what notes exist. This prevents errors from wrong paths or duplicate names.

### Example 1: User asks to read a note (check first)
User: "Show me my project plan"

Step 1 - Check what exists:
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree
```
Output:
```json
{"tree": [{"path": "Projects", "type": "dir"}, {"path": "Projects/Plan.md", "type": "file"}]}
```

Step 2 - Now read the correct path (always base64 encode):
```bash
# echo -n "Projects/Plan.md" | base64 → UHJvamVjdHMvUGxhbi5tZA==
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl read UHJvamVjdHMvUGxhbi5tZA== --base64
```
Output:
```json
{"path": "Projects/Plan.md", "content": "# Project Plan\n\n## Goals\n..."}
```

### Example 2: User asks to create a note (check first to avoid duplicates)
User: "Create a meeting notes file"

Step 1 - Check what already exists:
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree
```
Output:
```json
{"tree": [{"path": "Projects", "type": "dir"}, {"path": "Projects/Plan.md", "type": "file"}]}
```

Step 2 - No "Meeting Notes" exists, safe to create (do NOT duplicate title as heading):
```bash
# echo -n "Meeting Notes.md" | base64 → TWVldGluZyBOb3Rlcy5tZA==
# echo -n "## Agenda\n\n- Item 1\n- Item 2\n" | base64 → IyMgQWdlbmRhCgotIEl0ZW0gMQotIEl0ZW0gMgo=
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl create TWVldGluZyBOb3Rlcy5tZA== IyMgQWdlbmRhCgotIEl0ZW0gMQotIEl0ZW0gMgo= --base64
```
Output:
```json
{"status": "ok", "path": "Meeting Notes.md"}
```

### Example 3: User asks about vault contents
User: "What's in my notes?"

```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree --depth 2
```
Output:
```json
{"tree": [{"path": "Projects", "type": "dir"}, {"path": "Projects/Plan.md", "type": "file"}, {"path": "Ideas.md", "type": "file"}]}
```

Then summarize for user: "You have a Projects folder with Plan.md, and an Ideas.md file at the root."

### Example 4: Complex workflow with source and output notes
User: "According to the source note 'AI Digest Sources.md', browse the sources and output the digest to 'digest/2025-01-28-digest.md'"

Step 1 - Check what exists:
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl tree
```
Output:
```json
{"tree": [{"path": "AI Digest Sources.md", "type": "file"}, {"path": "digest", "type": "dir"}, {"path": "digest/2025-01-27-digest.md", "type": "file"}]}
```

Step 2 - Validate:
- Source "AI Digest Sources.md" exists
- Output "digest/2025-01-28-digest.md" does NOT exist, will use `create`

(If source didn't exist: STOP and ask user "I couldn't find 'AI Digest Sources.md'. Did you mean one of these: [list alternatives]?")

(If output already existed: use `append` instead of `create`)

Step 3 - Read the source note (always base64 encode):
```bash
# echo -n "AI Digest Sources.md" | base64 → QUkgRGlnZXN0IFNvdXJjZXMubWQ=
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl read QUkgRGlnZXN0IFNvdXJjZXMubWQ= --base64
```
Output:
```json
{"path": "AI Digest Sources.md", "content": "# AI Digest Sources\n\n- https://example.com/article1\n- https://example.com/article2\n"}
```

Step 4 - Browse sources and generate digest content (done by bot outside this skill)

Step 5 - Write output to vault (do NOT duplicate title as heading):
```bash
# echo -n "digest/2025-01-28-digest.md" | base64 → ZGlnZXN0LzIwMjUtMDEtMjgtZGlnZXN0Lm1k
# echo -n "## Summary\n\nKey points from today's sources...\n" | base64 → IyMgU3VtbWFyeQoKS2V5IHBvaW50cyBmcm9tIHRvZGF5J3Mgc291cmNlcy4uLgo=
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl create ZGlnZXN0LzIwMjUtMDEtMjgtZGlnZXN0Lm1k IyMgU3VtbWFyeQoKS2V5IHBvaW50cyBmcm9tIHRvZGF5J3Mgc291cmNlcy4uLgo= --base64
```

(If output already existed, use `append` instead:)
```bash
ssh -4 -p 2222 ${VAULT_SSH_USER}@localhost vaultctl append ZGlnZXN0LzIwMjUtMDEtMjgtZGlnZXN0Lm1k IyMgVXBkYXRlCi4uLg== --base64
```

Attribution

rondoflowrondoflow
View sourceMore from rondoflow →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

693621 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →