# FlowScope UX Flow Analyser (pypi · flowscope-mcp)

Turn YouTube product demos into structured UX flow reports, and compare demos side by side.

- Trust score: 59/100 (low)
- Change this week: −10
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-24

## Components

- pypi · `flowscope-mcp`: 59/100 (this document), [markdown](https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp.md), [page](https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp)

## Channel facts

- Registry: `pypi`
- Package: `flowscope-mcp`
- Version: `0.1.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-24.

- **Supply Chain Security**: 50/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - 1 of 28 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 17 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 82/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1204 tokens (~109/item across 11 items; 10 tools + 1 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 33/100
  - Stability observed for 10 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (90% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 12 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the FlowScope UX Flow Analyser MCP server?

FlowScope UX Flow Analyser runs locally as a PyPI package, launched with uvx flowscope-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add davidngugi-flowscope -- uvx flowscope-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "davidngugi-flowscope": {
      "command": "uvx",
      "args": [
        "flowscope-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "davidngugi-flowscope": {
      "command": "uvx",
      "args": [
        "flowscope-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add davidngugi-flowscope -- uvx flowscope-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "davidngugi-flowscope": {
      "type": "local",
      "command": [
        "uvx",
        "flowscope-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add davidngugi-flowscope --command uvx --arg flowscope-mcp
```

### Hermes

```yaml
mcp_servers:
  davidngugi-flowscope:
    command: "uvx"
    args: ["flowscope-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "davidngugi-flowscope": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "flowscope-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add davidngugi-flowscope -t stdio -c uvx -a flowscope-mcp
```

### Other

```json
{
  "mcpServers": {
    "davidngugi-flowscope": {
      "command": "uvx",
      "args": [
        "flowscope-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-23 (score 59, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-21 (score 58, +4)

- [functional improvement] Stability: unverified → 0.23

### 2026-10-18 (score 54, −15)

- [security regression] Malware scan: pass → unverified

### 2026-10-17 (score 69, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-14 (score 54, +15)

- [security regression] Malware scan: pass → unverified
- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] Schema quality: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional] First check of Tool coverage: 0
- [functional] First check of Schema quality: fail
- [functional] First check of Schema quality: fail
- [functional] First check of Destructive annotations: 100
- [functional] First check of Tool coverage: 90
- [functional] First check of Schema quality: good

### 2026-10-12 (score 39)

First indexed and scored.

## MCP tools (10)

### `flowscope_health_check` (~63 tokens)

Check FlowScope readiness

Check that the FlowScope backend is running and that its dependencies (ffmpeg, yt-dlp, an LLM provider key) are ready. Call this before submitting a video so a missing prerequisite is reported as a config problem rather than a failed multi-minute job.

Output parameters:

- `advice` (string): What to do about blocking_problems, if any.
- `backend_url` (string): Base URL this server is talking to.
- `blocking_problems` (array): Human-readable list of reasons analysis would fail; empty when ok is true.
- `ffmpeg_available` (boolean): ffmpeg present; required for frame extraction.
- `ffmpeg_version`: Detected ffmpeg version.
- `llm_comparison_model` (string): Model used for comparisons.
- `llm_provider` (string): Default provider.
- `llm_providers_available` (array): Providers with credentials, e.g. ['openai'].
- `llm_synthesis_model` (string): Model used to synthesise the flow.
- `llm_vision_model` (string): Model used for the frame (vision) stage.
- `ok` (boolean): True when the backend answered and can analyse videos.
- `yt_dlp_available` (boolean): yt-dlp importable; required to download videos.

### `flowscope_list_videos` (~66 tokens)

List analysed videos

List videos FlowScope has been asked to analyse, newest first, with the status of each one's latest job. Call this first: results are cached by YouTube video id, so an existing entry means the analysis is free to re-read and must not be resubmitted.

Output parameters:

- `count` (integer)
- `videos` (array)

### `flowscope_job_status` (~67 tokens)

Check analysis progress

Report the progress of a video's latest analysis job. Use this to poll after flowscope_analyze_video returns while still running, or to find out why an analysis failed. Poll every 20-30 seconds; do not resubmit the URL.

Input parameters:

- `video_id` (string, required)

Output parameters:

- `error_message`
- `is_terminal` (boolean): True when the job finished, successfully or not.
- `job_id`
- `next_action` (string): What the caller should do next.
- `percent_complete`: Null when the total is unknown.
- `progress_current` (integer)
- `progress_total` (integer)
- `stage_detail`
- `stage_label` (string): Human-readable version of status.
- `status` (string): queued|downloading|...|synthesizing|done|error
- `video_id` (string)

### `flowscope_get_report` (~84 tokens)

Read a video's UX report

Return the finished UX analysis of one video: the ordered user flow, synthesised UX insights, and per-screen findings with the visible UI elements and the narration that accompanies each screen. This is the primary tool for answering UX questions about a video.

Input parameters:

- `include_transcript` (boolean)
- `max_frames` (integer)
- `video_id` (string, required)

Output parameters:

- `channel`
- `duration_seconds`
- `flow_steps` (array): Ordered user flow.
- `frames` (array): Per-screen findings, in video order.
- `message` (string): Why the report is missing, when ok is false.
- `note` (string): Caveats about this report, if any.
- `ok` (boolean): False when the report could not be read.
- `report_markdown` (string): The whole report rendered as Markdown.
- `screen_count` (integer): Number of distinct screens analysed.
- `status`: Latest job status.
- `title`
- `transcript_excerpt`: Transcript text, present only when include_transcript was true.
- `transcript_source`
- `truncated` (boolean): True when frames were dropped to respect max_frames.
- `ux_insights` (array): Synthesised UX observations.
- `video_id` (string)
- `youtube_url` (string)

### `flowscope_compare_videos` (~95 tokens)

Compare the UX of several videos

Compare two or more already-analysed videos and return their common patterns, their divergences, and a per-stage matrix. Requires at least two videos that have finished analysing. The comparison is an LLM call, so it takes a while and costs money; results are cached until one of the videos is re-analysed.

Input parameters:

- `force_refresh` (boolean)
- `video_ids` (array, required)

Output parameters:

- `common_patterns` (array): What the products share.
- `divergences` (array): Where they differ.
- `estimated_cost_usd`
- `message` (string): Why the comparison failed, when ok is false.
- `ok` (boolean): False when the comparison failed.
- `report_markdown` (string): The comparison rendered as Markdown.
- `stage_matrix` (array): Per-stage, per-product breakdown.
- `video_ids` (array)

### `flowscope_analyze_video` (~164 tokens)

Analyse a YouTube product demo

Submit one or more YouTube URLs for UX analysis. FlowScope downloads each video, transcribes it, extracts and deduplicates screenshots of distinct screens, describes each screen with a vision model, and synthesises the flow. This is slow (minutes per video) and the vision steps cost money, so check flowscope_list_videos first -- an already-analysed video is reused for free and immediately. By default this waits for completion and returns the reports; if the analysis needs longer than wait_seconds it returns status 'running' with video ids to poll.

Input parameters:

- `force` (boolean)
- `force_local_transcription` (boolean)
- `urls` (array, required)
- `wait` (boolean)
- `wait_seconds` (number)

Output parameters:

- `message` (string): Plain-language summary of what happened.
- `reports` (array): Filled in only when every video finished within the wait budget.
- `reused_count` (integer)
- `started_count` (integer)
- `status` (string): 'done', 'reused', 'running' or 'error'.
- `videos` (array)

### `flowscope_retry_video` (~64 tokens)

Retry a failed analysis

Re-queue the last analysis job for a video after a failure, or start a fresh job when the previous one finished. Stages that already completed are reused, so a retry does not re-download or re-bill them.

Input parameters:

- `video_id` (string, required)

Output parameters:

- `action` (string)
- `job_id`
- `message` (string)
- `ok` (boolean)
- `status`
- `video_id` (string)

### `flowscope_reanalyze_video` (~74 tokens)

Redo the derived analysis for a video

Discard a video's frames, per-screen analyses and synthesised flow, then redo extraction and analysis from the downloaded video. Use this after changing the LLM model or prompt. It deletes stored results and spends money again. Any comparison that included this video is dropped too.

Input parameters:

- `video_id` (string, required)

Output parameters:

- `action` (string)
- `job_id`
- `message` (string)
- `ok` (boolean)
- `status`
- `video_id` (string)

### `flowscope_delete_video` (~65 tokens)

Delete a video and its analysis

Permanently remove a video's analysis from FlowScope. With delete_media true it also deletes the downloaded video and extracted frames from disk, which cannot be undone. Ask the user before using delete_media.

Input parameters:

- `delete_media` (boolean)
- `video_id` (string, required)

Output parameters:

- `action` (string)
- `job_id`
- `message` (string)
- `ok` (boolean)
- `status`
- `video_id` (string)

### `flowscope_frame_image` (~86 tokens)

View a screen screenshot

Return the actual screenshot for one screen of a video, as an image the caller can look at. Use this when the written report is not enough and the visual layout matters, for example to judge spacing, hierarchy or visual style. Takes the video id and the frame_id from flowscope_get_report.

Input parameters:

- `frame_id` (string, required)
- `video_id` (string, required)

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp#diagnostics

## Score history

- 2026-10-24: 59
- 2026-10-23: 59
- 2026-10-22: 58
- 2026-10-21: 58
- 2026-10-20: 54
- 2026-10-19: 54
- 2026-10-18: 54
- 2026-10-17: 69
- 2026-10-16: 54
- 2026-10-15: 54
- 2026-10-14: 54
- 2026-10-13: 39
- 2026-10-12: 39

## Common questions

### What is the FlowScope UX Flow Analyser MCP server?

FlowScope UX Flow Analyser is an MCP server listed in the public MCP registry as io.github.DavidNgugi/flowscope. Turn YouTube product demos into structured UX flow reports, and compare demos side by side. This page covers its PyPI package (flowscope-mcp).

### Is the FlowScope UX Flow Analyser MCP server safe to use?

FlowScope UX Flow Analyser scores 59 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 24 October 2026. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the FlowScope UX Flow Analyser MCP server expose?

FlowScope UX Flow Analyser exposes 10 tools: flowscope_health_check, flowscope_list_videos, flowscope_job_status, flowscope_get_report, flowscope_compare_videos, and 5 more. Their descriptions and schemas cost roughly 828 tokens of context every time the server is loaded.

### Is the FlowScope UX Flow Analyser MCP server still maintained?

FlowScope UX Flow Analyser is still listed as active in the MCP registry. We last reached this channel on 24 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the FlowScope UX Flow Analyser MCP server under?

FlowScope UX Flow Analyser declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/flowscope-mcp/
- Socket report: https://socket.dev/pypi/package/flowscope-mcp
- Repository: https://github.com/DavidNgugi/flowscope
- Changelog RSS feed: https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp.json
- HTML version of this page: https://verifymcp.io/servers/davidngugi-flowscope/flowscope-mcp