Agent skill
ssh-vault
Execute SSH commands on remote hosts via SSH Vault MCP. Use when: user asks to run commands on vault-managed hosts, or when in a Discord channel mapped to a vault host. NOT for: hosts with direct SSH access or openclaw nodes. Ask user for their vault URL if not known. Requires env vars: SSH_VAULT_URL (vault endpoint), SSH_VAULT_AGENT_PRIVATE_KEY and SSH_VAULT_AGENT_PUBLIC_KEY (Ed25519 agent keypair for request signing). The private key is used only for signing vault API requests — never reuse keys tied to other services.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/ssh-vault
SKILL.md
SSH Vault
Execute SSH commands on remote hosts through a self-hosted SSH Vault instance.
User has deployed SSH Vault via Docker. Ask for their vault URL if not known.
Required env vars:
SSH_VAULT_URL— vault URLSSH_VAULT_AGENT_PRIVATE_KEY— agent Ed25519 private key (base64)SSH_VAULT_AGENT_PUBLIC_KEY— agent Ed25519 public key (base64) Fingerprint is auto-derived from public key.
Docker Setup (user-managed)
docker run -d -p 3001:3001 \
-v vault-data:/app/data \
-v vault-config:/app/config \
qsobad/ssh-vault-mcp:latest
/app/config/config.yml— auto-created with localhost defaults if missing/app/data— encrypted vault storage- Set `SSH_VAULT_ORIGIN env var for custom origin (e.g. https://ssh.example.com))
Execution
node scripts/vault.mjs exec <host> <command> [timeout]
Happy path (has session): returns { stdout, stderr, exitCode } immediately.
No session: returns needsApproval:
{ "needsApproval": true, "approvalUrl": "...", "listenUrl": "...", "execRequestId": "..." }
Approval Flow
- Show
approvalUrlto user — opens/approve-execpage showing host + command - User authenticates with Master Password + Passkey → vault unlocks, command executes
- Listen on
listenUrl(SSE) for result:data: {"status":"pending"} data: {"status":"approved"} data: {"status":"executing"} data: {"status":"completed","stdout":"...","stderr":"...","exitCode":0,"sessionId":"..."} - Save
sessionIdto/tmp/ssh-vault-session.json— subsequent commands skip approval
Shell metacharacters (&&, ;, |, $(), backticks) are all allowed in commands.
Other Commands
node scripts/vault.mjs status # Vault lock status
node scripts/vault.mjs session # Cached session info
node scripts/vault.mjs register # Register agent
node scripts/vault.mjs check-unlock <id> # Check unlock & get session
node scripts/vault.mjs hosts # List hosts (needs session)
Adding Hosts
Hosts can be added by agent via API — user provides credential (password/key) during approval:
node scripts/vault.mjs request-host <name> <ip> <user> [port] [authType]
Error Handling
needsApproval→ show approvalUrl, listen SSEHost not found→ check withhostsAgent not registered→ runregister
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?