ssh-mcp-server-copilot

Provides an MCP server (stdio transport) that lets VS Code Copilot Chat manage remote Linux hosts over SSH using natural-language queries mapped to a fixed set of template-based, policy-scoped tools (read-only, controlled mutation, and privileged operations requiring approval).

Evaluated Apr 04, 2026 (0d ago)
Homepage ↗ Repo ↗ Infrastructure mcp ssh vscode copilot security automation remote-operations python
⚙ Agent Friendliness
53
/ 100
Can an agent use this?
🔒 Security
77
/ 100
Is it safe for agents?
⚡ Reliability
26
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
70
Error Messages
0
Auth Simplicity
62
Rate Limits
0

🔒 Security

TLS Enforcement
85
Auth Strength
78
Scope Granularity
72
Dep. Hygiene
55
Secret Handling
90

Security posture is heavily guardrailed per README: no raw shell (template-only), regex-validated parameters, path traversal blocking, secret redaction in outputs, approval workflow for privileged Tier 2 operations with HMAC-verified one-time tokens, and tamper-evident hash-chained audit logging. TLS is not directly discussed (SSH transport typically encrypted, but docs focus on SSH, not HTTPS/TLS). Dependency hygiene is not evidenced in provided material (scores are estimate-based from dependency list, not from CVE scanning results). Rate limiting and explicit error-handling guidance are not described.

⚡ Reliability

Uptime/SLA
0
Version Stability
35
Breaking Changes
40
Error Recovery
30
AF Security Reliability

Best When

When you want an LLM/agent workflow to operate remote servers under strict guardrails (no raw shell, parameter validation, path policies, and auditable/approvable privileged actions).

Avoid When

When you cannot restrict templates and roles appropriately, or when operational procedures for approvals and audit review are not feasible.

Use Cases

  • Read-only operational queries on remote Linux hosts (host discovery, system facts, audit log viewing).
  • Controlled execution of whitelisted command templates (e.g., disk usage, log tails) with regex-validated parameters.
  • File transfer and SFTP operations constrained by allowed paths/extensions and download justification requirements.
  • Privileged SSH key and certificate lifecycle actions gated by an approval workflow and HMAC-verified one-time tokens.

Not For

  • Arbitrary remote command execution or interactive shell access.
  • Environments where approval workflows and audit logging cannot be enforced/operationally supported.
  • Use as a general-purpose API server for non-SSH workflows (it is specialized for SSH/MCP tool operations).

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: SSH key/ssh-agent usage for SSH connectivity Optional bearer token via SSH_MCP_AUTH_TOKEN (not clearly production-grade in docs) Two-party approval workflow for Tier 2 (request/approve/consume approval tokens)
OAuth: No Scopes: No

The docs describe role-based access (developer/operator/admin/auditor) and a Tier 2 approval flow with HMAC-verified one-time approval tokens. It also mentions SSH_MCP_AUTH_TOKEN for bearer-token auth (empty = dev mode), but does not clearly document how roles map to that token or how scopes are enforced for Tier 0/1.

Pricing

Free tier: No
Requires CC: No

No commercial pricing information provided; package appears open-source (MIT) based on provided metadata.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • MCP tools only work when Copilot Chat is switched to "Agent" mode (explicitly called out).
  • Long-running work uses background jobs and polling/cancellation; agents may need guidance to choose sync vs background tools.
  • Approval tokens appear one-time/consumed; agents must not retry Tier 2 tool calls without re-issuing approval if token is expired/consumed.
  • Template-only execution means agents must select/fit requests to existing templates/regex constraints; otherwise operations may fail due to validation/path policy restrictions.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for ssh-mcp-server-copilot.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered