Compound Labs MCP
NPM · KYNTH-MCP · SCANNED SEP 20
Compound Labs: eleven keyless lookups. Model pricing, dep health, stack cost, registries.
Available components
Deprecated
This server is marked deprecated in the MCP registry.
The registry records this reason: The studio is Compound Labs. This server is published as tech.thecompound/compound-mcp, npm compound-mcp.
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. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 39 of 123 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency45
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 15 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability72
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 1751 tokens (~159/item across 11 items; 11 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 11 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 11 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
How do I install the Compound Labs MCP server?
Compound Labs MCP runs locally as an npm package, launched with npx -y kynth-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · kynth-mcp
claude mcp add studio-kynth-kynth-mcp -- npx -y kynth-mcp
{
"mcpServers": {
"studio-kynth-kynth-mcp": {
"command": "npx",
"args": [
"-y",
"kynth-mcp"
]
}
}
} {
"servers": {
"studio-kynth-kynth-mcp": {
"command": "npx",
"args": [
"-y",
"kynth-mcp"
]
}
}
} codex mcp add studio-kynth-kynth-mcp -- npx -y kynth-mcp
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"studio-kynth-kynth-mcp": {
"type": "local",
"command": [
"npx",
"-y",
"kynth-mcp"
],
"enabled": true
}
}
} openclaw mcp add studio-kynth-kynth-mcp --command npx --arg -y --arg kynth-mcp
mcp_servers:
studio-kynth-kynth-mcp:
command: "npx"
args: ["-y", "kynth-mcp"] {
"McpServers": {
"studio-kynth-kynth-mcp": {
"Transport": "stdio",
"Command": "npx",
"Arguments": [
"-y",
"kynth-mcp"
]
}
}
} assistant mcp add studio-kynth-kynth-mcp -t stdio -c npx -a -y kynth-mcp
{
"mcpServers": {
"studio-kynth-kynth-mcp": {
"command": "npx",
"args": [
"-y",
"kynth-mcp"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 20 Sept 26 0
- Stability: 0.97 → pass security
- 19 Sept 26 +6
- Stability: 0.93 → unverified ▼ security
- Tool safety: pass → unverified ▼ security
- Source repository: fail → pass ▲ security
- Capabilities: pass → unverified ▼ functional
- Tool coverage: 100 → unverified ▼ functional
- First check of Schema quality: unverified functional
- Package version: 0.3.0 → 0.4.0 functional
- 17 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
- 15 Sept 26 +2
- CVE-2026-82562 no longer affects this package ▲ security
- CVE-2026-82417 no longer affects this package ▲ security
- Known CVEs: fail → pass ▲ security
- 14 Sept 26 −2
- Stability: pass → 0.80 functional
- 12 Sept 26 0
- Stability: 0.97 → pass security
- 10 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 8 Sept 26 −1
- Stability: pass → 0.87 functional
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 20 Sept 2026 · Analysed npm/kynth-mcp@0.4.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 123 packages
| Packages resolved | 123 |
|---|---|
| Stale | 39 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
check_project_maintenance Check whether a developer tool is still maintained ~114
Read StillShipping, which tracks whether developer tools and libraries are still actively shipping — last release, last commit, and whether the project reads as dead. Use before recommending a dependency. A library that was healthy at training time may have been abandoned since, and this is the check that catches it.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | number | – | How many to return, 1-10. Default 5. |
| q | string | – | Tool or library name to look up. Omit to list recently-declared-dead projects. |
No output schema declared.
No examples provided.
compare_ai_models Compare current AI model pricing and rankings ~163
Read ToolDrift, which tracks AI model pricing and which models the major AI-powered apps actually route to, sampled nightly. Use when asked which model to use, what a model costs now, or what changed recently — model pricing moves faster than any training cutoff, so answering from memory is reliably wrong.
| Name | Type | Req | Description |
|---|---|---|---|
| board | string | – | Which ranking to read, as "scope/key". Budget bands: budget/free, budget/value, budget/balanced, budget/premium. Per-tool routing: tool/claude-code, tool/codex-cli, tool/cline, tool/aider, tool/roo-c… |
| limit | number | – | How many models to return, 1-10. Default 10. |
No output schema declared.
No examples provided.
compare_app_builders Compare AI app builders on whether their output clears App Store review ~228
Read StoreReady, a sourced comparison of the AI mobile-app builders (Rork, Vibecode, a0.dev, Replit, Bolt and the rest) on one question: can the app it hands you go up on your own Apple Developer account and survive review. Returns what each builder outputs, whether you can export the source, who submits the binary, the App Store guidelines in play, and a verdict of ships / ships-with-caveats / blocked / unknown. Use when asked which AI app builder to pick, or whether one can actually ship to the App Store — the answer changes as builders launch and change, and every verdict here is attached to a source that was read on a stated date.
| Name | Type | Req | Description |
|---|---|---|---|
| builder | string | – | One builder slug or name for the full record with its evidence, e.g. "rork", "vibecode", "bolt". |
| limit | number | – | How many to return, 1-10. Default 10. |
| verdict | string | – | Filter the list: ships, ships-with-caveats, blocked, unknown. |
No output schema declared.
No examples provided.
estimate_stack_cost Estimate the monthly bill for a named stack at a given scale ~130
Price a whole stack at a chosen user count using StackTab's verified plan data. Returns the plan each service lands on and an itemised bill — which lines are included in the base fee and which are metered overages. Use when someone asks "what will this cost at N users" instead of estimating from memory.
| Name | Type | Req | Description |
|---|---|---|---|
| stack | string | yes | Comma-separated service slugs, e.g. "supabase,clerk,polar,vercel". Look them up with lookup_service_pricing. |
| users | number | – | Monthly active users to price at. Default 10,000. |
No output schema declared.
No examples provided.
grade_starter_kit Look up a graded SaaS/Next.js starter kit ~105
Read KitGrade, which installs and grades public starter kits and boilerplates — what is actually wired up (auth, billing, email, tests), what only appears in the README, and how fresh the dependencies are. Use when someone asks which starter to build on.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | number | – | How many to return, 1-10. Default 5. |
| q | string | – | Kit name to look up. Omit for the top-graded kits. |
No output schema declared.
No examples provided.
lookup_ada_report Look up an ADA Title II website accessibility report ~145
Look up a US local-government (.gov) domain in the CivicBinder Municipal Web Accessibility Index. Returns the accessibility grade (A-F), WCAG 2.1 AA violation counts from an axe-core scan, the failing rules, the entity's ADA Title II compliance deadline (April 2027 or April 2028), and a link to the full public report page. Use for questions like "is cityofx.gov ADA compliant" or "how accessible is this county website".
| Name | Type | Req | Description |
|---|---|---|---|
| domain | string | yes | The .gov domain to look up, e.g. "denvergov.org" or "cityofmadison.com". Protocol and paths are stripped automatically. |
No output schema declared.
No examples provided.
lookup_nonprofit_status Look up nonprofit IRS / California good standing by EIN ~132
Check a US nonprofit's standing by EIN against the IRS auto-revocation list (tax-exempt status revoked for three consecutive missed Form 990 filings) and California Registry of Charities delinquency/suspension lists. Returns revocation and reinstatement dates, whether the streamlined 15-month reinstatement window is still open, and whether AB 488 requires fundraising platforms to block donations. A "clear" result means the EIN is on none of the tracked lists.
| Name | Type | Req | Description |
|---|---|---|---|
| ein | string | yes | The organization's 9-digit EIN, with or without a dash, e.g. "12-3456789". |
No output schema declared.
No examples provided.
lookup_service_pricing Look up current published pricing for a SaaS service ~142
Read StackTab, which re-checks the published pricing of the services a product runs on (auth, database, hosting, payments) and records when each plan was last verified against the vendor's own page. Use when asked what something costs or what its free tier includes. Prices and free-tier limits change often enough that a remembered figure is a guess; each plan here carries the date it was last checked.
| Name | Type | Req | Description |
|---|---|---|---|
| category | string | – | Filter by category: "auth", "database", "hosting" or "payments". |
| service | string | – | A specific service slug or name, e.g. "clerk", "neon", "stripe". |
No output schema declared.
No examples provided.
search_agent_configs Search real AGENTS.md / CLAUDE.md files from public repositories ~249
Search RuleStack, an index of real agent instruction files (AGENTS.md, CLAUDE.md, Cursor rules, Copilot instructions, Windsurf, GEMINI.md, Cline) harvested from public GitHub repositories and scored for quality. Use when writing or reviewing an agent config and you want to see how well-maintained projects with a given stack actually write theirs — "show me CLAUDE.md files from Next.js repos", "what do good AGENTS.md files put in their build section".
| Name | Type | Req | Description |
|---|---|---|---|
| format | string | – | Comma-separated format slugs: agents-md, claude-md, cursor-rules, copilot-instructions, windsurf-rules, gemini-md, cline-rules. |
| limit | number | – | How many to return, 1-10. Default 5. |
| min_quality | number | – | Minimum quality score, 0-100. |
| stack | string | – | Comma-separated stack slugs, ANDed. e.g. "nextjs,typescript". |
| tag | string | – | Comma-separated section tags, ANDed: build, test, security, code-style, architecture, git-pr, do-not, docs. |
No output schema declared.
No examples provided.
search_agent_skills Search published Claude Code skills, plugins and marketplaces ~155
Search SkillWorks, an index of publicly published agent skills, plugins and skill marketplaces, ranked by installs and repository signal. Use before writing a skill from scratch to check whether one already exists, or when a user asks "is there a skill for X" / "what should I install for Y". Returns the install command for each result, so the answer is directly actionable.
| Name | Type | Req | Description |
|---|---|---|---|
| kind | string | – | Filter to one of: skill, subagent, plugin, marketplace. |
| limit | number | – | How many to return, 1-10. Default 5. |
| q | string | yes | What the skill should do, in plain words. e.g. "pdf generation", "code review", "terraform". |
No output schema declared.
No examples provided.
search_component_registries Search shadcn-style component registries ~188
Search BlockDex, a cross-registry index of shadcn-compatible components, blocks, hooks and libraries from dozens of public registries. Use when a user wants a UI component and you would otherwise hand-write it — "is there a date range picker block", "which registries have a kanban board". Returns the registry, the item type and its dependencies so you can judge the cost of pulling it in.
| Name | Type | Req | Description |
|---|---|---|---|
| access | string | – | Filter by whether the item is free to install: free, paid, unknown. |
| kind | string | – | Filter by item kind: component, block, hook, example, lib, theme, other. |
| limit | number | – | How many to return, 1-10. Default 5. |
| q | string | yes | What the component does. e.g. "data table", "auth form", "kanban". |
No output schema declared.
No examples provided.
What is the Compound Labs MCP server?
Compound Labs MCP is listed in the public MCP registry as studio.kynth/kynth-mcp. Compound Labs: eleven keyless lookups. Model pricing, dep health, stack cost, registries. This page covers its npm package (kynth-mcp).
Is the Compound Labs MCP server safe to use?
Compound Labs MCP scores 83 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. It declares no install or post-install scripts. 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 Compound Labs MCP server expose?
Compound Labs MCP exposes 11 tools: lookup_ada_report, lookup_nonprofit_status, search_agent_configs, search_agent_skills, search_component_registries, and 6 more. Their descriptions and schemas cost roughly 1,751 tokens of context every time the server is loaded.
Is the Compound Labs MCP server still maintained?
Compound Labs MCP is marked deprecated in the MCP registry. We last reached this channel on 20 September 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 Compound Labs MCP server under?
Compound Labs MCP declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.