app-soul-mcp
Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
weberwang
Source: github_code
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
read_imageRead a single image from the mood board directory and return it directly to you (the calling multimodal AI) for analysis. Use this after download_mood_board to extract palette, visual style, metaphors, and atmosphere from a reference image.
read_mood_board_dirRead all images from OUTPUT_DIR/mood_board/ and return them directly to you (the calling multimodal AI) for analysis. Returns all images in a single response so you can synthesize a unified style profile: palette, visual metaphors, atmosphere.
get_brand_guide_promptReturns the brand guide JSON schema and a ready-to-use prompt. Call this first, then use the prompt yourself to generate the brand guide — no internal AI call needed.
save_brand_guideSave a brand guide JSON to a local file so it can be referenced by later steps (palette extraction, code generation, etc.).
load_brand_guideLoad a previously saved brand guide JSON from disk.
get_prototype_promptReturns a ready-to-use prompt for generating a functional prototype. Tech stack is caller-defined — not limited to HTML/CSS. Use this prompt yourself to generate the prototype — no nested AI call. Then pass the output to save_code_files.
get_final_code_promptReturns a ready-to-use prompt for generating production code from brand guide + palette. Use this prompt yourself to generate the code — no nested AI call. Then pass the output to save_code_files.
save_code_filesParse <code_file name="...">...</code_file> blocks from AI-generated code and save each file to disk. Also accepts legacy <file> blocks for compatibility. Use after get_prototype_prompt or get_final_code_prompt generation.
get_design_spec_promptReturns a ready-to-use prompt for generating a comprehensive UI/UX design specification. The spec covers semantic color tokens (light + dark), typography scale, spacing system, component dimensions, motion guidelines, state patterns, and accessibility requirements — everything needed to guide implem…
save_design_specSave a UI/UX design specification JSON to disk so it can be loaded by get_final_code_prompt as implementation guidance.
load_design_specLoad a previously saved design spec from disk. Use the result as the designSpec input to get_final_code_prompt.
download_mood_boardDownload mood board reference images to a local directory from a list of URLs. URLs are typically extracted by an external browser/crawler MCP (e.g. Playwright MCP) browsing Dribbble, Mobbin, Figma Community, etc.
extract_paletteExtract a harmonious color palette from a reference image in OUTPUT_DIR/mood_board/ using median-cut quantization. Each color is assigned a semantic role (background, surface, accent, text).
extract_palette_from_dirExtract and merge palettes from all images in OUTPUT_DIR/mood_board/. Returns a deduplicated representative palette across all reference images.
review_ui_screenshotRead a rendered UI screenshot and return it directly to the calling multimodal AI with a structured review prompt. Use this after generating and rendering code to critique visual quality, brand alignment, hierarchy, density, and anti-pattern drift.
review_ui_screenshot_dirRead all screenshots from a directory and return them to the calling multimodal AI with a structured review prompt for flow-level comparison. Use this for multi-screen review after rendering a prototype or app flow.
save_ui_reviewSave a structured UI review JSON result to disk so it can be reused in later design or code refinement steps.
load_ui_reviewLoad a previously saved UI review JSON from disk.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
sharp inherited vulnerabilities in libvips: CVE-2026-33327, CVE-2026-33328, CVE-2026-35590, CVE-2026-35591
Tool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
18/18 tools missing one or more hints — read_image (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); read_mood_board_dir (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); get_brand_guide_prompt (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +15 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Descriptions match behaviour
1 tool describes read intent but its handler mutates — download_mood_board (line 81: fs.mkdir(targetDir, { recursive: true }))
Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.
Tool inputs are validated
Only 7/18 tool handlers declare input schemas (39%)
Declare an inputSchema with zod/joi/yup on every tool definition.
License file
No license file
Add a LICENSE file (MIT, Apache-2.0, etc.).
Tests exist
No test files found
Add tests that exercise each declared tool.
Production dependencies are patched
0 critical, 1 high severity in production deps — sharp@0.33.5 (high)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
Dependency freshness
1/5 production deps stale: node-fetch@2023-11-30 (2.8y)
Tool description accuracy
download_mood_board: description implies read-only but handler writes/deletes/executes
Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/weberwang/app-soul-mcp)?variant=verified from the URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check