slideless-mcpofficial
F
licenseNot graded
qualityD
maintenanceMCP server that wraps the Slideless HTTP API as tools for listing, sharing, uploading, and managing HTML presentations from any MCP host without installing the CLI.
-
htmldrop-mcp
Publish HTML or Markdown content to a public hosted URL and manage those published sites via MCP.
htmldrop_publish – Publish HTML or Markdown as a hosted page; returns the public share URL, site id, and slug. Optionally set a custom slug and title.
htmldrop_list – List all sites the authenticated tenant has published, with each site's slug, id, and public URL.
htmldrop_delete – Delete a published site by its id; the public URL immediately returns 404.
Works from MCP clients (Claude Code, Claude Desktop, Cursor, Cline, etc.) via a remote OAuth server or a local stdio server with an API token.
Enforces htmldrop plan limits and surfaces clear errors when caps are reached.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@htmldrop-mcpPublish a single-page HTML CV for Alex Rivera."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@htmldrop.app/mcp
Model Context Protocol server for htmldrop — publish HTML or Markdown to a real hosted URL straight from Claude Code, Claude Desktop, Cursor, Cline, or any MCP-aware client.
Note: this is htmldrop.app. An unrelated project also called "htmldrop" exists at htmldrop.link (npm
htmldrop-mcp) — different product, different API.
Quick start — remote server (recommended)
No install, no API keys. Add the hosted server and sign in with OAuth in your browser:
claude mcp add --transport http htmldrop https://htmldrop.app/mcpThe first time the model publishes, your browser opens a sign-in — approve once and you're done. Works the same in any MCP client that supports remote (streamable HTTP) servers with OAuth; see the per-client pages on htmldrop.app/agents.
Then just ask: "Make a single-page HTML CV for Alex Rivera and publish it to htmldrop." The model calls htmldrop_publish and replies with the live URL.
Related MCP server: publish-artifacts-mcp
Interactive clients — always use the remote
Claude Code (above), Claude Desktop (Settings → Connectors → Add custom connector → https://htmldrop.app/mcp), Cursor (Settings → MCP → Add remote server), and other remote-capable clients should all use the hosted URL. You sign in once in the browser; no credential is ever written to a config file. Per-client walkthroughs: htmldrop.app/agents.
CI, scripts, and headless use — local server (stdio)
Automation can't do a browser sign-in, so the stdio server authenticates with an API token — created at htmldrop.app/dashboard/settings → API tokens (hsk_live_…, shown once). Store it in your secret manager or CI's encrypted secrets and reference it — never write the literal value into a config file or repository.
GitHub Actions example:
- name: Publish report page
env:
HTMLDROP_API_TOKEN: ${{ secrets.HTMLDROP_API_TOKEN }}
run: |
npx -y @htmldrop.app/mcp & # stdio server for your MCP-driving tool
# …or call the REST API directly: https://docs.htmldrop.app/api/authenticationShell/session example (token from your OS keychain or secret manager, not typed inline):
export HTMLDROP_API_TOKEN="$(security find-generic-password -s htmldrop -w)" # macOS Keychain
npx -y @htmldrop.app/mcpIf a client genuinely can't speak remote MCP and isn't automation (rare now), the classic mcpServers JSON block with an env entry works — accept that it stores the token in that client's config file, and prefer the .mcpb bundle for Claude Desktop, which keeps the token in the OS keychain instead.
Tools
Tool | What it does |
| Publish an HTML or Markdown document. Returns the public share URL, site id, and slug. Accepts an optional |
| List every site the authenticated tenant owns. |
| Delete a site by id. |
Configuration (local server)
Env var | Default | Purpose |
| (required for stdio) | The |
|
| Override for staging or self-hosted htmldrop instances. |
| (derived from | User-facing origin for share links ( |
Plan limits
The MCP server hits the same API your dashboard does, so plan caps apply — free accounts keep 3 drops, paid plans (from $5/mo) raise caps and add permanence, password protection, version history, and custom domains. Current numbers: htmldrop.app/#pricing. When you hit a cap the tool call returns a clear error (plan_limit) that the model surfaces in chat — no silent failure.
Security notes
The remote server (
https://htmldrop.app/mcp) follows the MCP authorization spec: OAuth 2.1 with PKCE, discovery via RFC 9728 protected-resource metadata, and short-lived tokens your client refreshes automatically. Tokens are never passed in URLs. This is the recommended path for humans in MCP clients — no static credential exists anywhere.Local-server API tokens (
hsk_live_…) carry full tenant privileges — treat them like a password. Keep them out of source control, store them in a secret manager or your CI's encrypted secrets (not a committed config file), and revoke from Settings → API tokens the moment one may have leaked (revocation is effective immediately). The.mcpbone-click install for Claude Desktop stores the token in your OS keychain rather than a plaintext file.The stdio server runs on your machine. No traffic flows through htmldrop other than the API calls the tool makes on your behalf.
Development
npm ci
npm test # builds + runs the node:test suiteThis repository is a read-only mirror of the mcp/ directory in htmldrop's main (private) repository — the source of truth. Issues and PRs are welcome here; accepted changes are applied upstream and sync back automatically.
License
MIT — © Middle East Software Solutions Limited
Available Tools
3 toolshtmldrop_deleteADestructiveIdempotent
Delete a site by its id. The id is the opaque string returned from htmldrop_list (NOT the slug). Once deleted the public URL returns 404 immediately.
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | The site's id, as returned by htmldrop_list. |
Output Schema
ParametersJSON Schema
| Name | Required | Description |
|---|---|---|
| id | Yes | The deleted site's id. |
| deleted | Yes | Always true on success. |
TDQS
A4.5/5.0
Behavior4/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and idempotentHint=true, so the agent knows it's a destructive, repeatable operation. The description adds a concrete behavioral detail: 'Once deleted the public URL returns 404 immediately,' which informs the agent of the immediate, observable effect. This goes beyond the annotation metadata without contradicting it.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences long, with the primary action stated first and the clarifying detail second. Every word adds value: the id-source clarification prevents misuse, and the 404 note sets expectations. No fluff or repetition.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness5/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is simple (one parameter), has annotations covering safety (destructive/idempotent) and an output schema, so the description doesn't need to explain return values. It covers the only ambiguity (which id to use) and adds the immediate 404 behavior. No critical information is missing for an agent to invoke this tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters4/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema covers the single parameter completely (100% coverage), giving the baseline of 3. The description enhances this by explaining that the id is an 'opaque string' and explicitly warning against using the slug, which is not present in the schema's description. This added semantic distinction justifies a score above the baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose5/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description opens with a specific verb and object: 'Delete a site by its id.' This clearly distinguishes it from sibling tools like htmldrop_list (which lists sites) and htmldrop_publish (which publishes). It further clarifies that the id is an opaque string from htmldrop_list, not the slug, removing ambiguity about the resource identifier.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines4/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for when to use this tool: when you need to delete a site. It also gives practical guidance on which id to pass (the opaque string from htmldrop_list, not the slug). It doesn't explicitly name alternatives, but the action is unambiguous and the id-source clarification is a strong usage cue.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
htmldrop_listARead-onlyIdempotent
List every site the authenticated tenant has published. Returns each site's slug, id, and public URL.
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Output Schema
ParametersJSON Schema
| Name | Required | Description |
|---|---|---|
| sites | Yes | Every published site for the authenticated tenant. |
TDQS
A4.2/5.0
Behavior4/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint safely. The description adds valuable context by specifying the 'authenticated tenant' scope and the returned fields (slug, id, public URL), which goes beyond the annotations without contradicting them.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
A single, front-loaded sentence communicates the purpose and return values without any wasted words. It is appropriately sized for a tool with no inputs.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness4/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity, no parameters, and an existing output schema likely listing the response structure, the description is sufficient. It covers the key behavioral aspect (listing every published site) and indicates the return fields, though it could mention pagination or rate limits if applicable.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters4/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, so the baseline is 4. The description correctly does not need to explain any parameters, and the schema provides full coverage with an empty definition.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose5/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function with a specific verb ('List'), resource ('site'), and scope ('every site the authenticated tenant has published'). It naturally differentiates from sibling tools htmldrop_publish and htmldrop_delete, which perform write operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines3/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies when to use the tool (to view published sites) but does not explicitly mention alternatives or exclusion criteria. Sibling tools exist but are not referenced, so the agent must infer that this is the read-only counterpart.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
htmldrop_publishA
Publish an HTML or Markdown document as a hosted page on htmldrop. Returns the public share URL. Use this whenever a user asks to publish, share, or 'put on the web' some HTML / Markdown content. Pass exactly one of html or markdown.
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| html | No | Raw HTML document content. Either this or `markdown` must be set. | |
| slug | No | Optional URL slug (3-63 chars, lowercase letters, numbers, hyphens). If omitted, a random slug is assigned. | |
| title | No | Optional title shown in the dashboard list. | |
| markdown | No | Markdown content. Will be rendered to HTML on the server. Either this or `html` must be set. |
Output Schema
ParametersJSON Schema
| Name | Required | Description |
|---|---|---|
| id | Yes | Opaque site id. Pass to htmldrop_delete to remove later. |
| url | Yes | Public share URL of the published page. |
| slug | Yes | Resolved slug for the published site. |
TDQS
A4.3/5.0
Behavior3/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already disclose that this is a non-read-only, open-world, non-idempotent operation. The description adds that it creates a hosted page and returns a share URL, but these are largely implied by the annotations and the operation type. No additional behavioral traits like auth requirements or rate limits are mentioned. No contradiction exists.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
The description consists of two concise sentences. The first sentence states the primary action and result, the second gives usage trigger and the key parameter constraint. Every sentence earns its place with no filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness5/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With a fully documented input schema, an output schema present, and annotations covering mutation, external side effects, and non-idempotency, the description needs only to clarify usage and the exclusive parameter choice. It does both effectively, making the tool's invocation well-specified for an agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters4/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already provides detailed descriptions for all four parameters. The description adds the valuable cross-parameter constraint: 'Pass exactly one of html or markdown.' This clarifies mutual exclusivity, which is not fully explicit in the schema (each property only states 'Either this or markdown must be set'). Thus, the description meaningfully supplements the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose5/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function: 'Publish an HTML or Markdown document as a hosted page on htmldrop.' It specifies the resource, the action, and the outcome ('Returns the public share URL'). This distinguishes it from sibling tools (list/delete).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines4/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides explicit when-to-use guidance: 'Use this whenever a user asks to publish, share, or put on the web some HTML / Markdown content.' This offers clear context and trigger phrases. However, it does not explicitly state when not to use it or mention alternative tools, so it's a 4 rather than a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
3 tool updates
v0.3.4- First observed
htmldrop_delete - First observed
htmldrop_list - First observed
htmldrop_publish
TDQS
A4.4/5.0
Disambiguation5/5
Each tool has a distinct action: publish, list, and delete. There is no overlap or ambiguity between the tools.
Naming Consistency5/5
All tool names follow the same pattern: htmldrop_ + verb (publish, list, delete). This is perfectly consistent and predictable.
Tool Count5/5
Three tools is well-scoped for a focused service that publishes, lists, and deletes hosted HTML/Markdown. Each tool is necessary and none are redundant.
Completeness4/5
The core lifecycle is covered: publish (create), list (read), and delete. A minor gap is the lack of an update or edit capability, but the main workflows are complete.
Maintenance
ActivityMaintained
ResponsivenessNo issues
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Publish HTML, Markdown, and multi-file sites as shareable URLs instantly via MCP.
Publish HTML or Markdown from any MCP client to a live, shareable URL on htmldrop.app.
Publish and manage existing HTML presentations from an MCP-capable Agent.
List, share, upload, and manage Slideless HTML presentations from any MCP host.
Related MCP Servers
- FlicenseNot gradedqualityCmaintenanceEnables publishing, updating, and sharing HTML artifacts with strict security isolation (origin separation, CSP, API keys) via MCP tools.1-
- AlicenseNot gradedqualityCmaintenancePublish and manage shareable HTML/Markdown pages with access control and comments via MCP clients.MIT
@dropthis/mcpofficial
AlicenseAqualityAmaintenancePublish content (HTML, files, or URLs) and get a permanent public URL from any MCP-compatible agent, with local and remote connection options.29244MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/middleeastsoftware/htmldrop-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server