Skip to main content
Glama
Sign In
swayammaheshwari

docstar-mcp

by swayammaheshwari
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

docstar-mcp

Remote MCP server (Streamable HTTP) that lets AI clients work with a DocStar workspace. It runs as a Cloudflare Worker and is a thin layer over the DocStar API (hitman-api); it never touches the database. Users sign in with OAuth 2.0, so no API token is needed, and the user's existing DocStar permissions apply to everything the server does.

Tools

Tool

What it does

list_collections

Collections the signed-in user can access

list_pages

Flat page list of a collection (use parentId for the tree)

get_page

A page with its content as Markdown

search_docs

Semantic search in a collection

create_page

New draft page from Markdown

update_page

Change name and/or content (Markdown); page goes back to draft

publish_page

Publish a page (its parent must already be published)

Related MCP server: Autonomous Documentation MCP

Authentication

The server uses OAuth 2.0. The first time a client connects, it opens a browser window where you sign in to DocStar and approve access. There are no API keys to create or paste, and no credentials are stored in this repository.

Connect from Claude Code

claude mcp add --transport http docstar-workspace https://mcp.docstar.io/mcp

Then run /mcp inside Claude Code and choose docstar-workspace to complete the browser sign-in.

Connect from Cline

Open MCP Servers → Configure MCP Servers and add this to cline_mcp_settings.json:

{
  "mcpServers": {
    "docstar": {
      "type": "streamableHttp",
      "url": "https://mcp.docstar.io/mcp",
      "disabled": false,
      "autoApprove": []
    }
  }
}

type must be streamableHttp (the server has no SSE endpoint). Cline opens the DocStar sign-in in your browser on first connect. Agent-driven install steps are in llms-install.md.

Security

  • Sign-in uses OAuth 2.0; the server never sees your DocStar password.

  • The server is a thin proxy over the DocStar API and never touches the database.

  • Your DocStar permissions decide what each tool can read or change.

  • Nothing is stored in this repository; no secrets are required to use the hosted server.

Run locally

npm install
DOCSTAR_API_URL=http://localhost:2000 npm run dev   # http://localhost:3100/mcp

The local server also needs the organization: send an X-Docstar-Org-Id header or set DOCSTAR_ORG_ID.

MCP Registry

The server is listed in the official MCP Registry as io.github.swayammaheshwari/docstar-mcp; server.json is the listing. The logo shown there comes from icons in server.json (an HTTPS URL; the server also reports it in its MCP serverInfo). To publish a change, bump version in server.json, then:

mcp-publisher login github  
mcp-publisher publish

Roadmap

  1. OAuth scopes (finer-grained, read-only vs write access)

  2. create_collection, resources (docstar://page/{id}), prompts, higher-level workflow tools

License

MIT - see LICENSE.

Available Tools

3 tools
apply_doc_updateC

Apply generated documentation to a file

NameRequiredDescriptionDefault
contentYesThe documentation content
file_pathYesPath of the file to update/create

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full behavioral burden. It implies a mutation ('apply... to a file') but does not state whether existing content is overwritten or appended, whether the file/directories are created, whether it is reversible, or what is returned.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single front-loaded sentence with no waste. It is appropriately terse, though arguably under-specified rather than merely concise.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a mutation tool with no annotations and no output schema, the definition should clarify the effect on existing file content and the return value. With 100% schema coverage the parameters are covered, but the behavioral picture an agent needs before writing to a file is largely absent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so both parameters (file_path, content) are already documented in the schema, setting the baseline at 3. The description adds no format, path-resolution, or content-shape details beyond what the schema provides.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (apply), resource (generated documentation) and target (a file), so the agent knows it writes documentation into a file. However, it does not differentiate itself from siblings like generate_docs_for_changes, leaving the boundary between generating and applying implicit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is no explicit when-to-use guidance, no prerequisites, and no mention of the sibling tools or how they chain (e.g., generate then apply). Usage is only vaguely inferable from the name.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

generate_docs_for_changesC

Generate documentation for changes

NameRequiredDescriptionDefault
context_filesNoList of file paths to read for extra context
changes_summaryYesSummary of changes (diff)

TDQS

C2.3/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It does not state whether documentation is written to disk, returned as text, requires permissions, or what side effects occur; it only implies that something is generated.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single short phrase is concise, but it is under-specified rather than well-structured. There is no front-loading because there is nothing substantive to front-load.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with no annotations, no output schema, and two inputs, the description is far too thin. It omits usage context, behavioral traits, and output expectations, leaving the agent with only the name and input schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, with both parameters documented (changes_summary and context_files). The description adds no parameter semantics, but the baseline is 3 when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose2/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description restates the tool name ('generate_docs_for_changes' → 'Generate documentation for changes') without adding scope or distinguishing it from siblings like get_recent_changes or apply_doc_update. An agent learns nothing beyond the name.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No when-to-use guidance, no prerequisites, and no mention of alternatives. The two siblings (get_recent_changes, apply_doc_update) are never referenced, leaving the agent to guess the workflow.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_recent_changesC

Get recent git changes

NameRequiredDescriptionDefault
limitNoNumber of commits to check
include_diffNoWhether to include the diff content

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so the description carries the full burden and does not meet it. It does not disclose scope ('recent' = how far back?), whether diffs are included, whether it is read-only, or what the return shape looks like. Only the word 'git' hints at behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single four-word sentence, front-loaded and undeniably concise. It is arguably too terse to carry meaning, but as conciseness goes it wastes nothing.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a git-inspection tool with no annotations and no output schema, the description should explain the retrieval scope and whether it is safe/read-only. It leaves both unaddressed, and gives no hint about the sibling relationship that would make this tool's role obvious.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents both parameters (limit, include_diff) with defaults. The description adds nothing about parameters, which is acceptable given full schema coverage — baseline 3.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clear verb+resource ('Get recent git changes') and the tool is distinguishable from siblings by inspection (generate_docs_for_changes and apply_doc_update are about docs, this is about retrieving changes). However the description doesn't explicitly differentiate itself from siblings, and 'recent' is undefined without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No when-to-use guidance, no alternatives named, no conditions. An agent must infer that this is the retrieval step that feeds generate_docs_for_changes, but that relationship is never stated.

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.

  1. 3 tool updatesv0.1.0
    • First observedapply_doc_update
    • First observedgenerate_docs_for_changes
    • First observedget_recent_changes
Deploy Server
Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

TDQS

B3.2/5.0

Scored across 3 tools

Disambiguation5/5

The three tools form a clear sequential pipeline: retrieve changes, generate docs, apply update. Each has a distinct action and resource, so an agent can easily select the right one.

Naming Consistency5/5

All names use snake_case with a verb-first pattern (get_, generate_, apply_) followed by a noun/object. Consistent and predictable throughout.

Tool Count5/5

Three tools perfectly match the narrow purpose of a documentation-generation pipeline; each earns its place with no redundancy.

Completeness4/5

The surface covers the end-to-end workflow (fetch, generate, apply) but lacks optional steps like previewing generated docs, targeting specific files, or rolling back changes.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers