Interactive Feedback
SAFEgrade B · trust 84/100Enhanced MCP server for interactive user feedback and command execution in AI-assisted development, featuring dual interface support (Web UI and Desktop Application) with intelligent environment detection and cross-platform compatibility.
Overview
From the repository's own README, as read at the audited commit.
# MCP Feedback Enhanced**🌐 Language / 語言切換:** **English** | [繁體中文](README.zh-TW.md) | [简体中文](README.zh-CN.md)**Original Author:** [Fábio Ferreira](https://x.com/fabiomlferreira) | [Original Project](https://github.com/noopstudios/interactive-feedback-mcp) ⭐**Enhanced Fork:** [Minidoracat](https://github.com/Minidoracat)**UI Design Reference:** [sanshao85/mcp-feedback-collector](https://github.com/sanshao85/mcp-feedback-collector)> ## 📢 Maintenance Status (2026-08)>> The project is maintained again. **Please upgrade to v2.6.1** — it fixes a command execution vulnerability:>> ```bash> uvx mcp-feedback-enhanced@latest> ```>> **What changed in v2.6.1:**> - 🔒 **Command execution removed** — fixes [#219](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/219) (unauthenticated WebSocket could execute arbitrary programs). The old blocklist only caught shell metacharacters, but since execution used `shell=False` metacharacters were never the risk — `cat`, `curl`, `wget`, `python` passed straight through, and auto-command was **enabled by default**. The feature is gone for good. See [SECURITY.md](SECURITY.md).> - 🔒 **Cross-Site WebSocket Hijacking fixed** (reported privately as `GHSA-cmr5-gpm3-79vf`, `GHSA-2wx7-r4rh-f663`): browsers are not restricted by the same-origin policy when opening a WebSocket, so a malicious page could make your browser connect to the local `/ws`. `Origin` is now validated before `accept()`, and cross-origin attempts are rejected with 403.> - 🐛 Fixed the Starlette breaking change that made the Web UI return 500 ([#213](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/213), [#217](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/217), [#221](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/221), [#228](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/228)).> - 🐛 Fixed image serialization ([#154](https://github.com/Minidoracat/mcp-feedback-enhanced/issues/154) and related) by switching to standard `m
Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add mcp-feedback-enhanced-desktop -- uvx mcp-feedback-enhanced-desktop
{
"mcpServers": {
"mcp-feedback-enhanced-desktop": {
"command": "uvx",
"args": [
"mcp-feedback-enhanced-desktop"
]
}
}
}Exposed tools (1) 1 read · 0 write · 0 destructive
| Tool | Risk | Description |
|---|---|---|
get_system_info | read |
Details
- Source
- minidoracat/mcp-feedback-enhanced
- PyPI
mcp-feedback-enhanced-desktop- License
- NOASSERTION
- Stars
- 3,768 · pushed 7d ago
Trust audit
Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (5 observation(s))
- Network
- declared (5 observation(s))
- Shell
- declared (1 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (25)
icon.icns
.bumpversion.cfg
.pre-commit-config.yaml
allow_origins=["http://127.0.0.1:*", "http://localhost:*"],
const backendUrl = 'http://127.0.0.1:8765';
await app.launch_tauri_app("http://127.0.0.1:8765")await app.launch_tauri_app("http://127.0.0.1:8765")await app.launch_tauri_app("http://127.0.0.1:8765")<link rel="icon" type="image/x-icon" href="data:image/x-icon;base64,AAABAAEAEBAAAAEAIABoBAAAFgAAACgAAAAQAAAAIAAAAAEAIAAAAAAAAAQAABILAAASCwAAAAAAAAAAAAD///8A////AP///wD///8A////AP///wD///8A////AP///wD/
image_bytes = base64.b64decode(img["data"], validate=True)
image_bytes = base64.b64decode(img["data"])
PNG_BYTES = base64.b64decode(
assert base64.b64decode(result[0].data) == PNG_BYTES
assert base64.b64decode(result[0].data) == PNG_BYTES
sudo apt-get update
sudo apt-get install -y \
if ! sudo apt-get install -y libayatana-appindicator3-dev; then
sudo apt-get install -y libappindicator3-dev
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
- `0.0.0.0`: Binds all interfaces. ⚠️ **Not recommended**: the Web UI and `/ws` endpoint have **no authentication**, so anyone who can reach the port can read session content (including project paths
> the Web UI and `/ws` endpoint have no authentication, so binding publicly lets anyone on the network read your session and submit feedback.
src/mcp_feedback_enhanced/desktop_release/mcp-feedback-enhanced-desktop-linux
src/mcp_feedback_enhanced/desktop_release/mcp-feedback-enhanced-desktop-macos-arm64
src/mcp_feedback_enhanced/desktop_release/mcp-feedback-enhanced-desktop-macos-intel
src/mcp_feedback_enhanced/desktop_release/mcp-feedback-enhanced-desktop.exe
Gates applied: no_behavioural_pass.
Audited 2026-09-14 · audit v0.4.0 · source sha e990a29086d8 · full audit: observations/trust-audit/mcp-server/minidoracat__interactive-feedback-4.json · Report an issue or request a re-scan
Audit history
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-09-14 | e990a29086d8 | SAFE | B | 84 | first audit |
Also audited
Servers with a similar trust profile.
Questions
What is the Interactive Feedback MCP server?
Enhanced MCP server for interactive user feedback and command execution in AI-assisted development, featuring dual interface support (Web UI and Desktop Application) with intelligent environment detection and cross-platform compatibility.
What tools does Interactive Feedback expose?
1 in total: 1 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Interactive Feedback safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (84/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Interactive Feedback need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (e990a29086d8), read on 2026-09-14. The repository is watched and re-audited when it changes.
Provenance: OBSERVED · read 2026-09-14 · job trust-audit-2026-09-14