Back to search
MCPScanner
MCP serverGitHub
MCP Auditor 🥕
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/abhinav-prabhakar--mcpscanner)HTML
<a href="/listings/abhinav-prabhakar--mcpscanner"><img src="/api/listings/abhinav-prabhakar--mcpscanner/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
# Go 1.26+ go install github.com/Abhinav-Prabhakar/MCPScanner/cmd/mcpscanner@latest mcpscanner # discover and audit local MCP client configs mcpscanner --demo # audit the deliberately hostile demo server
Or build from source:
git clone https://github.com/Abhinav-Prabhakar/MCPScanner.git cd MCPScanner && make build
---
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 245 lines
- <div align="center">
- # MCPScanner
- **Audit an MCP server before you hand it your files, your credentials and your agent.**
- `74 rules` · `27 permission classes` · `source + runtime` · `zero tool calls`
- [](https://go.dev)
- [](LICENSE)
- [Quick start](#quick-start) · [Discovery](#one-command-every-client) · [Drift](#catch-the-rug-pull) · [Rules](#rules-not-vibes) · [CLI](#cli-reference)
- </div>
- 
- MCP servers decide what your model can do. MCPScanner reads the source, performs a read-only handshake, maps every capability, and shows you the blast radius before you connect.
- <div align="center">
- <img src=".github/assets/demo.gif" alt="MCPScanner scanning the hostile demo server">
- <br>
- <sub><b>The built-in hostile demo.</b> 13 tools, 5 resources, 2 prompts, 38 findings, score 100. No child process, no socket.</sub>
- </div>
- ---
- ## Quick start
- ```bash
- # Go 1.26+
- go install github.com/Abhinav-Prabhakar/MCPScanner/cmd/mcpscanner@latest
- mcpscanner # discover and audit local MCP client configs
- mcpscanner --demo # audit the deliberately hostile demo server
- ```
- Or build from source:
- ```bash
- git clone https://github.com/Abhinav-Prabhakar/MCPScanner.git
- cd MCPScanner && make build
- ```
- ---
- ## One command. Every client.
- Run `mcpscanner` with no arguments. It finds Claude Desktop, Claude Code, Cursor, VS Code, Windsurf and Cline configs, then lets you pick what to audit. Selection only reads JSON; nothing is executed until you approve the probe.
- 
- | Merged config scan | Server identity checks |
- |:---:|:---:|
- |  |  |
- | <sub>Six configs, 13 servers, one correlated report.</sub> | <sub>Lookalike names and duplicate identities across configs.</sub> |
- Dedicated paths still work exactly as before:
- ```bash
- mcpscanner ~/.cursor/mcp.json
- mcpscanner ./mcp.json
- ```
- ---
- ## Catch the rug pull.
- A server can pass review, then change its tool descriptions, schemas or annotations after you approve it. MCPScanner fingerprints the approved surface and turns every later change into a finding.
- ```bash
- mcpscanner --demo --save-baseline approved.json
- mcpscanner --demo --baseline approved.json --fail-on-drift
- ```
- <div align="center">
- <img src=".github/assets/drift.gif" alt="Baseline drift detection in the MCPScanner TUI">
- <br>
- <sub><b>Rug-pull detection.</b> Approve once. Diff on every scan. Exit non-zero in CI on drift.</sub>
- </div>
- | Drift summary | What changed, line by line |
- |:---:|:---:|
- |  |  |
- | <sub>Seven changes since the approved baseline.</sub> | <sub>Description, schema, annotation, capability, resource and tool changes — with evidence.</sub> |
- Tracked surfaces: tool descriptions, input schemas, safety annotations, server version and protocol, capabilities, resources and prompts.
- ---
- ## What it sees
- <table>
- <tr>
- <td width="50%"><img src=".github/assets/permissions.png" alt="Permission map"><br><sub><b>27 permission classes.</b> Execution, filesystem, secrets, network, data, host and model reach — with the tools behind each.</sub></td>
- <td width="50%"><img src=".github/assets/tools.png" alt="Tool inspector"><br><sub><b>Tool-level reach.</b> Names, descriptions, schemas, defaults and annotations, scored and cross-checked.</sub></td>
- </tr>
- <tr>
- <td><img src=".github/assets/findings.png" alt="Findings list"><br><sub><b>Findings with proof.</b> Severity, CWE, source location, runtime evidence and remediation.</sub></td>
- <td><img src=".github/assets/source.png" alt="Source analysis"><br><sub><b>The code behind the server.</b> Languages, entrypoints, dependencies, env vars and endpoints.</sub></td>
- </tr>
- </table>
- **Correlation is the point.** Credential access **+** outbound requests becomes exfiltration. Shell execution **+** injection text becomes remote instruction execution. Browser profiles **+** network becomes session theft.
- **Authentication posture** is classified from config, `WWW-Authenticate` and public OAuth metadata: `none`, API key, bearer, basic, browser sign-in or OAuth. PKCE support and plaintext token endpoints are checked; no credential is ever sent.
- **Dangerous parameter names** are flagged too — a tool that asks for the system prompt, conversation history, SSH keys, browser cookies or the process environment is not a normal task input.
- ---
- ## Reports, not screenshots you have to trust
- 
- One report, four encodings, stable exit codes:
- ```bash
- mcpscanner ./server --no-tui # terminal report
- mcpscanner ./server --format json --save report.json # machine-readable
- mcpscanner ./server --format markdown --save report.md # PR-friendly
- mcpscanner ./server --format sarif --save report.sarif # code scanning
- mcpscanner ./server --fail-on high # exit 2 on high+
- mcpscanner ./server --baseline approved.json --fail-on-drift
- ```
- ---
- ## Rules, not vibes
- **74 rules** across **15 categories**, each with severity, CWE and remediation.
- `Secrets 11` · `Drift 7` · `Execution 7` · `Filesystem 6` · `Isolation 6` · `Transport 6` · `Deception 5` · `Prompt injection 5` · `Supply chain 4` · `Authentication 3` · `Exposure 3` · `Hygiene 3` · `Network 3` · `Privacy 3` · `Identity 2`
- 
- ```bash
- mcpscanner rules # full catalogue with remediation
- mcpscanner rules --json # the same data for tooling
- ```
- ---
- ## Safety model
- 
- - **Remote probes are read-only.** Only `initialize` and the list methods are sent. No tool, resource read or prompt is ever invoked.
- - **Local code requires consent.** stdio servers are never spawned until you approve them or pass `--yes`.
- - **The execution guard is on by default.** Shell wrappers, metacharacters, inline evaluation and dangerous binaries are refused.
- - **No dangerous client capabilities.** No sampling, elicitation or roots are advertised.
- - **Secrets and terminal output are sanitized.** Tokens are redacted; invisible unicode is stripped but reported as evidence.
- The scanner cannot prove a server is safe. It can tell you exactly what the server is asking for and what the code behind it can do.
- ---
- ## CLI reference
- <details>
- <summary><b>All flags and subcommands</b></summary>
- <br>
- | Flag | Description |
- |---|---|
- | `--no-probe` | Static analysis only. Never contact or execute the target. |
- | `--yes` | Consent to launch local stdio servers without the prompt. |
- | `--allow-shell` | **Dangerous.** Bypass the execution guard. |
- | `--json` | Shortcut for `--format json`. |
- | `--format` | `text`, `json`, `markdown` or `sarif`. |
- | `--save FILE` | Write the report to a file; format inferred from the extension. |
- | `--fail-on SEV` | Exit 2 when a finding at or above this severity exists. |
- | `--baseline FILE` | Compare the runtime surface against an approved baseline. |
- | `--save-baseline FILE` | Write the current runtime surface as a baseline. |
- | `--fail-on-drift` | Exit 2 when the baseline comparison finds changes. |
- | `--header K:V` | HTTP header for remote probes (repeatable). |
- | `--env K=V` | Environment override for stdio probes (repeatable). |
- | `--cmd CMD` | Launch command to probe alongside a source tree. |
- | `--transport` | Force `streamable-http`, `sse` or `stdio`. |
- | `--timeout` | Per-target probe timeout (default `25s`). |
- | `--insecure` | Skip TLS verification for HTTPS probes. |
- | `--no-tui` / `--tui` | Force the plain report or the interactive UI. |
- | `--demo` | Scan the embedded hostile demo. |
- | `--verbose` | Print warnings and probe trace details. |
- Subcommands: `rules`, `serve-demo --http :8765`, `version`, `help`.
- </details>
- ---
- ## Architecture
- ```text
- cmd/mcpscanner CLI, discovery, consent flow, exit codes
- internal/model shared report and auth types
- internal/rules rule catalogue + permission taxonomy
- internal/scan detection, config parsing, static analysis,
- stdio / Streamable HTTP / SSE transports,
- auth discovery, surface inspection, drift, scoring
- internal/tui Bubble Tea interface
- internal/report text, JSON, Markdown and SARIF renderers
- internal/demoserver embedded hostile MCP server
- internal/ui theme, gauges, panels, layout primitives
- demo/tapes VHS recordings used for every visual above
- ```
- Dependencies are deliberately small: Bubble Tea, Lip Gloss, Bubbles and `golang.org/x/term`. The JSON-RPC client is custom so every request, timeout and capability is visible.
- <details>
- <summary><b>Limitations</b></summary>
- <br>
- - Tools are never called, so real side effects and authorization cannot be proven from metadata.
- - Server-side isolation, rate limits and audit logging are out of scope.
- - Static analysis is pattern-based; minified or generated code reduces coverage.
- - Package signatures are not verified. Pin versions and read install scripts.
- - v0.1.0 is prototype-grade; interfaces and rule IDs may change before v1.0.
- </details>
- ---
- ## Development
- ```bash
- make check # gofmt + vet + tests + build
- make test
- make demo # build and launch the hostile demo
- make record # regenerate every GIF and screenshot with VHS
- ```
- See [SECURITY.md](SECURITY.md) for disclosure and the execution model.
- ## License
- MIT © 2026 Abhinav Prabhakar
- <br>
- <div align="center">
- <sub>Built for the moment before you click “connect”.</sub>
- </div>
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.