Jevcore MCP vs Vnsh
A side-by-side look at the Jevcore MCP and Vnsh MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Jevcore MCP for developer tools work as a local process; choose Vnsh for developer tools as a local process. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Jevcore MCP when
- You need dedicated capabilities in the Developer Tools domain.
- You prefer a local stdio subprocess transport.
- You want tools like
jev_ask, jev_rank, jev_check.
Choose Vnsh when
- You need dedicated capabilities in the Developer Tools domain.
- You prefer a local stdio subprocess transport.
- You want tools like
vnsh_read, vnsh_share, vnsh_share_file.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Developer Tools | Developer Tools |
| Quality signal | 67/100 · Great | 64/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | Not declared |
| Pricing | Free | Not declared |
| Required env vars | None listed | None listed |
| Verified owner | Community listing | Community listing |
| GitHub stars | 102 | 158 |
| Directory activity | 1 views 0 installs 0 upvotes | 1 views 0 installs 0 upvotes |
| Tools | 3 | 14 |
| Full listing | View Jevcore MCP | View Vnsh |
Tools & capabilities
Jevcore MCP tools 3
jev_askAsk TypeSafe Jev one or more typed questions about a single piece of state. Jev does not generate text: it returns a selected option and calibrated probabilities. Use it for judgments the rest of the work branches on — routing, classifying, scoring, triaging. Do NOT use it to write prose, explain, summarize, or generate code. Question types: "noul" is yes/no and returns the probability of true; "choice" picks one of the criteria keys you declare; "score" places the state on an ordered scale whose levels you declare in ascending order. A noul may also declare "boundary" — what true means and what false means — which is worth supplying whenever the line between them is not obvious. Batch related questions into one call: they are answered against the same state in one round-trip. A score question answers with a numeric "score" that may fall between levels, a "legend" mapping each level index to its description, and probabilities per level. This tool returns probabilities, not decisions — apply your own confidence threshold before acting, and treat a low-confidence answer as unknown rather than picking for it.jev_rankRank a list of candidates by how well each satisfies one stated criterion. Returns every candidate with an independent relevance probability, sorted. Use it to order search hits, triage a backlog, or find the one item that answers a question. The probabilities are independent per-candidate judgments, NOT a distribution that sums to 1. A flat set of scores means "none of these stands out", not a fine-grained ordering. What you send is capped, and the two caps behave differently. The candidate list is truncated at 16,000 characters, and the reply carries `truncated: true` when that happens — check for it, because a cut list is still ranked and the scores then describe what Jev actually read rather than what you sent. The generated questions are refused at 4,000 characters instead: exceeding that gives you an error asking for fewer or smaller items, never a silent best-N, so retrying the same call cannot help. Because the criterion is repeated into every candidate's question, that second cap is what bounds the candidate count — measured, an ordinary criterion fits 20 candidates and a 400-character criterion fits 7, while candidate length does not change the count at all and only starts truncating the list past roughly 1,600 characters per candidate.jev_checkCheck whether a piece of evidence supports a specific claim. Returns "supported", "contradicted", "conflicted", "insufficient", or "unknown", with the underlying probabilities. The distinction between "not supported" and "contradicted" is the reason to use this rather than reading the evidence yourself. It judges one claim against the evidence you give it. It does not search for evidence, and it cannot detect that you omitted the decisive passage.
Vnsh tools 14
vnsh_readReads any vnsh URL: decrypts legacy vnsh.dev/v links locally, or directly fetches a public vnshcontent.dev/p link. Public content is intentionally unencrypted.vnsh_shareEncrypts text content locally and uploads it to vnsh, returning a shareable URL. Use this when output exceeds ~50 lines or ~2000 characters, when generating complete code files, or when the user wants portable/shareable output. ALWAYS prefer vnsh_share over dumping large content into chat. The content is encrypted before upload - the server only sees encrypted bytes.vnsh_share_fileEncrypts a local file and uploads it to vnsh, returning a shareable URL. Use this for images, screenshots, PDFs, binaries, or any file that should not be loaded into context. More efficient than vnsh_share for large or binary files.vnsh_artifact_createCreates a permanent, service-readable Account Artifact in the signed-in user’s Library. Use this for durable knowledge that the user wants their other Agents to discover and update. Requires VNSH_TOKEN. This is different from an encrypted Incognito workspace: vnsh can read Account Artifact content so authorized Agents can collaborate on it.vnsh_artifact_listLists permanent Account Artifacts available to the connected user. Use this instead of asking the human to paste a link when continuing earlier account knowledge. Requires VNSH_TOKEN.vnsh_artifact_readReads the current content and version of one permanent Account Artifact. Pass the returned version to vnsh_artifact_update so another Agent cannot be overwritten.vnsh_artifact_updateCreates a new immutable version of an Account Artifact. Requires the base_version returned by vnsh_artifact_read. On conflict, re-read, merge, and retry; never silently overwrite.vnsh_workspace_createCreates a shared workspace and returns one link that any agent can read AND write. Use this when work will continue somewhere else: handing a plan, status doc, design, or investigation to another agent or another session, or when the user says they will pick this up in a different tool. Unlike vnsh_share (a one-shot snapshot), a workspace keeps the same URL as its content evolves. Content is encrypted locally; the server never sees the key. Anonymous workspaces expire 24h after the last write (ttl can extend this to a week). With VNSH_TOKEN, new workspaces and artifacts are kept until the account owner deletes them.vnsh_workspace_renewExtends the expiry of a vnsh workspace without changing its content. Use this when the user says a link is about to expire, or asks to keep something alive longer. Needs the edit link (a #w= fragment); a view-only #r= link cannot renew. The version is not bumped, so any pending edit still applies.vnsh_workspace_readReads the current content of a workspace: encrypted /w/ links with #w= or #r=, and public /p/ links with no fragment. Use this whenever the user provides such a link — it is how you pick up work another agent left for you. Returns the content and its version number; pass that version to vnsh_workspace_update to write safely.vnsh_workspace_updateReplaces the content of a vnsh workspace, keeping the same URL. Use this to record progress so far, revise a shared plan, or leave the next agent an updated document. Writes are version-checked: if someone else wrote first, this returns their current content so you can merge it with yours and call this again.vnsh_workspace_openDecrypts a vnsh workspace to a local temp file and opens it in the browser. Use this when the user wants to LOOK at a workspace rather than have its content read into context — HTML reports, dashboards, and diagrams render properly this way. Rendering happens locally from file://, so nothing is exposed to any server.- +2 more on the listing page
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Jevcore MCP
{
"mcpServers": {
"jevcore-mcp": {
"command": "npx",
"args": [
"-y",
"jevcore-mcp"
]
}
}
}
Vnsh
{
"mcpServers": {
"vnsh": {
"command": "npx",
"args": [
"-y",
"vnsh-mcp@1.9.0"
]
}
}
}
Frequently asked questions
Jevcore MCP is categorized under Developer Tools and uses a local stdio subprocess. Vnsh belongs to Developer Tools and uses a local stdio subprocess. Pick Jevcore MCP for developer tools work and Vnsh for developer tools.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client.
Jevcore MCP is listed as free. Vnsh has no declared pricing — check the project's own site. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add jevcore-mcp and vnsh as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Jevcore MCP has a quality signal of 67/100 (Great) with 1 views, 0 installs and 102 GitHub stars. Vnsh has 64/100 (Good) with 1 views, 0 installs and 158 GitHub stars.