How do I use anchor-mcp?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@anchor-mcp what contradictions exist in checkout-redesign?" 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.

anchor-mcp

by amitboker

Overview Schema Source Related Servers Score Discussions Admin

TypeScript

Hybrid

MCP Node License

Anchor reads a team's interviews, tickets and docs and builds a model of the product — the decisions, the contradictions, and everything nobody has answered. This is that model, exposed to your agent over the Model Context Protocol.

Six tools. All six return the fragments an answer came from, because an answer you cannot check is a guess with better formatting.

Quickstart

Runs with no account, no key and no network, against a bundled sample project. Pick your client:

Claude Code

claude mcp add anchor -- npx -y github:amitboker/anchor-mcp

Claude Desktop — claude_desktop_config.json

{
  "mcpServers": {
    "anchor": {
      "command": "npx",
      "args": ["-y", "github:amitboker/anchor-mcp"]
    }
  }
}

Cursor — .cursor/mcp.json

{
  "mcpServers": {
    "anchor": {
      "command": "npx",
      "args": ["-y", "github:amitboker/anchor-mcp"]
    }
  }
}

VS Code

code --add-mcp '{"name":"anchor","command":"npx","args":["-y","github:amitboker/anchor-mcp"]}'

Restart the client afterwards — all of them read MCP config at boot.

Not on npm yet, so the specs above install from this repository. npx builds it on first run, which takes a few seconds once and is cached after that.

Related MCP server: cortex-brain

Tools

Tool	What it does
`ask_brain`	Ask a question about the product. Returns the fragments that bear on it, each with its source.
`list_contradictions`	Statements that cannot both be true, both sides quoted, and the question that resolves each.
`list_gaps`	Questions no source answers, who can answer them, and what they block.
`list_sources`	Everything the brain has read, and how many fragments came out of each.
`get_fragment`	Open one fragment by id — full text and where it sits in its source.
`draft`	A PRD section, user stories or acceptance criteria, every line traced to a fragment.

What comes back

$ list_contradictions

3 contradictions in checkout-redesign.

## Cart retention window  (c-01)

A. The cart persists for 30 days on an unauthenticated session.
   — prd-checkout-v3 · §3.1 (PRD — Checkout redesign, v3)

B. Unauthenticated carts must expire within 7 days.
   — sec-review-q2 · Findings, item 4 (Security review — payments surface)

→ Does guest checkout keep a 30-day cart and require an identity, or drop
  to a 7-day window and stay anonymous?

Two documents, four months apart, in two different tools. Neither author was wrong. Nobody was going to notice.

Two things it will not do

It does not answer what it has not read:

$ ask_brain "quarterly revenue in Belgium"

Nothing in checkout-redesign answers that.

And it does not draft without sources:

$ draft --kind prd_section --topic "offline sync"

Not drafting "offline sync" — nothing in checkout-redesign supports it.

Both are the product working. A tool that fills a gap it cannot see the bottom of is the failure mode this exists to remove.

Sample data, and real projects

Out of the box the server runs on a bundled fixture — a checkout-redesign project with six sources and ten fragments. Every source, quote and person in it is invented, and every response says so in its footer.

Set ANCHOR_API_KEY and the same six tools read a real Project Brain instead. The output shape does not change; only the data does.

{
  "mcpServers": {
    "anchor": {
      "command": "npx",
      "args": ["-y", "github:amitboker/anchor-mcp"],
      "env": { "ANCHOR_API_KEY": "${ANCHOR_API_KEY}" }
    }
  }
}

Reference the variable from your shell rather than pasting a key into a file you might commit.

Status: preview. The tool surface is stable enough to build against. The hosted API behind ANCHOR_API_KEY is not open yet — until it is, every answer comes from the sample project.

Development

npm install
npm run build
npm run inspect

npm run inspect opens the MCP Inspector against your local build, which is the fastest way to see a tool's real output while changing it.

src/
  index.ts    server + tool registration
  tools.ts    tool bodies, callable without a transport
  brain.ts    the sample project
  cite.ts     citation formatting

One rule worth knowing before you add a tool: every response goes through answer() in cite.ts, and its cites argument is a non-empty tuple. An uncited answer is a type error rather than a matter of discipline — which is the only way a rule like that survives a deadline.

Logging goes to stderr, never stdout. stdout is the JSON-RPC transport, and one stray console.log on it corrupts the stream and takes the session with it.

License

MIT — see LICENSE.

Available Tools

6 tools

ask_brainAsk the Project BrainA

Ask a question about the product's own history — decisions, constraints, what customers said — and get the fragments that bear on it, each with the source it came from. Returns 'nothing answers that' rather than guessing when the brain has not read it.

ParametersJSON Schema

Name	Required	Description	Default
`limit`	No	Maximum fragments to return. Default 5.
`question`	Yes	A question about the product, in plain language.

TDQS

A4.4/5.0

Behavior4/5

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

No annotations are provided, so the description must disclose behavior on its own. It explicitly states the tool returns 'nothing answers that' rather than guessing when the brain has not read the source, and that results include the source. This covers the key behavioral trait of avoiding fabrication.

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, front-loaded with the action, and includes essential behavioral details without any redundant text.

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?

With no output schema and no annotations, the description covers the input scope, output format, and empty-result behavior. It could mention potential pagination or ordering details, but for a simple query tool with a limit parameter, it is sufficiently complete.

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 already documents both parameters with coverage 100%, so the baseline is 3. The description adds meaningful context for the 'question' parameter by specifying the kind of questions allowed (history, decisions, constraints, customer feedback), and clarifies the output format (fragments with sources), which enriches the parameter semantics.

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 uses a specific verb ('Ask') and identifies the resource ('the product's own history') and scope ('decisions, constraints, what customers said'). It distinguishes from sibling tools like get_fragment and list_sources by focusing on natural-language question retrieval rather than direct retrieval or enumeration.

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?

It gives clear context for when to use: asking plain-language questions about the product's history. It does not explicitly name alternatives or exclusions, but the context is unambiguous.

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

draftDraft from sourcesA

Draft a PRD section, user stories or acceptance criteria for a topic, with every line traced to the fragment behind it and the open questions listed underneath. Refuses to draft when no source supports the topic.

ParametersJSON Schema

Name	Required	Description	Default
`kind`	Yes	What to produce.
`topic`	Yes	The topic to draft, e.g. 'cart retention'.

TDQS

A4.4/5.0

Behavior5/5

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

With no annotations provided, the description carries full responsibility. It discloses key behaviors: every line traced to the source fragment, open questions listed below, and refusal when no supporting source. These go beyond the schema and give the agent an accurate mental model of the tool's operation.

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?

Two sentences, no filler. The first sentence immediately states the action and key output properties, and the second adds a decisive rejection condition. Every word earns its place.

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?

With no output schema, the description provides sufficient insight into the return structure (traced lines, open questions) and refusal case. It could have elaborated on where sources come from or how the draft is presented, but for a two-parameter tool, this is adequately complete.

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?

The input schema already describes both parameters (kind and topic) with 100% coverage, so the description adds little beyond restating the kinds in prose. It does not introduce new syntax or format details beyond what the schema provides, matching the baseline for high schema coverage.

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 uses a specific verb ('Draft') and names clear resources ('PRD section, user stories or acceptance criteria'). It also distinguishes itself from siblings by highlighting traceability to fragments and listing open questions, which none of the sibling tools imply.

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 conveys clear context: use this tool to draft from sources with traceability. It also gives an implicit when-not by stating it refuses when no source supports the topic, though it does not explicitly mention alternative tools or exclusion scenarios.

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

get_fragmentGet a source fragmentA

Fetch one fragment by id with its full text and where it sits in its source. Every other tool cites fragment ids; this is how you open one.

ParametersJSON Schema

Name	Required	Description	Default
`id`	Yes	A fragment id, e.g. f-04.

TDQS

A4/5.0

Behavior3/5

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

With no annotations, the description carries the full burden. It discloses that the tool returns full text and source position, which is useful, but it does not mention error handling (e.g., invalid id), potential empty results, or any permission requirements. For a read-only fetch, this is adequate but not rich.

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, front-loaded with the core function, and adds value with the sibling differentiation. There is no redundant or filler text.

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?

For a simple one-parameter fetch tool without an output schema, the description adequately covers purpose, return content, and usage context relative to siblings. It could optionally describe the exact response structure, but that is not essential given the tool's simplicity.

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?

The input schema provides 100% coverage for the single 'id' parameter, including an example. The description reinforces the parameter's role ('by id') but does not add further semantic details beyond what the schema already provides, so the baseline of 3 is appropriate.

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 a specific action ('Fetch one fragment by id') on a specific resource ('a source fragment') and adds what is returned (full text and location). It also distinguishes the tool from siblings by noting that other tools cite fragment ids and this is how to open one.

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 implies when to use the tool: when you have a fragment id and need to view its content or context. The phrase 'Every other tool cites fragment ids; this is how you open one' provides clear usage context, though it does not name a specific alternative or explicit 'when not to use' scenario.

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

list_contradictionsList contradictionsA

Statements in the project's sources that cannot both be true, with both sides quoted and the question somebody has to answer to resolve each one. This is usually the highest-value call: contradictions are what a person cannot hold in their head across six tools.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.5/5.0

Behavior4/5

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

With no annotations provided, the description takes on the full burden of explaining behavior. It reveals that the tool outputs quoted sides and a resolution question for each contradiction, giving a clear picture of return content. It does not mention side effects, but 'list' implies a read-only operation.

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, with the core functionality front-loaded in the first sentence and usage context in the second. No unnecessary detail is included, earning a top score.

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?

For a no-parameter tool, the description provides sufficient context: it states the inputs (project sources), outputs (quoted contradictions and resolution questions), and a prioritization hint. Despite lacking an output schema, the description covers the essential return structure, making it complete.

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 and the schema is empty (100% coverage), so parameter description is unnecessary. The baseline for zero parameters is 4, and the description correctly avoids fabricating parameter details.

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 defines the tool's purpose: listing contradictions from project sources, including quotes and resolution questions. It distinguishes itself from sibling tools like list_gaps and list_sources by focusing specifically on contradictions, making its role unambiguous.

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 statement 'This is usually the highest-value call' provides strong guidance on when to invoke this tool, suggesting it should be prioritized to surface conflicts. However, it does not explicitly mention alternatives or when not to use it, so it falls short of a 5.

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

list_gapsList open questionsA

Questions no source answers, with the role that can answer each and what it blocks. Deliberately returns no citations — there is nothing to cite, which is the point.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.2/5.0

Behavior4/5

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

With no annotations, the description carries the full burden. It explicitly discloses the deliberate absence of citations ('returns no citations — there is nothing to cite, which is the point') and describes the content of results (questions, answering role, blockers). This adds meaningful context, though it stops short of declaring read-only semantics or other operational details.

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, both concise and front-loaded. The first sentence states the core purpose, and the second explains a key behavioral nuance. Every sentence earns its place without unnecessary filler.

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?

For a zero-parameter tool with no output schema and no annotations, the description covers the essential details: what the tool returns (unanswered questions, responsible role, blockers) and a notable behavior (no citations). It could be more explicit about the output format (e.g., array of objects), but it is largely complete for simple list retrieval.

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 input schema is complete. Per the baseline for 0 params, a score of 4 is appropriate. The description adds no parameter-specific information because none exist, and none is needed.

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 states a specific verb and resource: 'Questions no source answers, with the role that can answer each and what it blocks.' This clearly explains what the tool does and distinguishes it from sibling list tools like list_sources and list_contradictions, which focus on different aspects.

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 the tool is used to find unanswered questions, but it does not explicitly state when to use it versus alternatives like list_contradictions or list_sources. No exclusions or alternative comparisons are provided, so usage guidance is only implicit.

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

list_sourcesList sourcesA

Everything the brain has read for this project — interviews, PRDs, tickets, threads, docs — with how many fragments came out of each. Use it to see what an answer could possibly be based on before you trust one.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.4/5.0

Behavior4/5

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

With no annotations, the description carries the full burden. It clearly implies a read-only listing operation and adds context about the content types and fragment counts. While it doesn't explicitly state 'read-only' or mention ordering/pagination, the behavior is transparent enough for a simple list tool.

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?

Two sentences: the first describes functionality, the second explains the use case. No fluff, perfectly front-loaded and concise.

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?

The tool is simple with no params or output schema, and the description covers what it returns and why to use it. It doesn't detail exact return shape or ordering, but that's acceptable for a listing tool with such clear intent.

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 adds value by explaining the output (fragment counts), which is informative but not parameter-related.

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 that the tool lists all sources the brain has read, with examples (interviews, PRDs, tickets, threads, docs) and fragment counts per source. This specific verb+resource is distinct from sibling tools like list_contradictions or list_gaps.

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 a clear usage context: 'Use it to see what an answer could possibly be based on before you trust one.' It does not explicitly exclude cases or name alternatives, but the intended use is evident.

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.

6 tool updatesv0.1.0Jul 31, 2026
- First observedask_brain
- First observeddraft
- First observedget_fragment
- First observedlist_contradictions
- First observedlist_gaps
- First observedlist_sources

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.3/5.0

Scored 2026-07-31 21:25 across 6 tools

Disambiguation5/5

Each tool serves a clearly distinct purpose: fetching a fragment, drafting content, querying the brain, listing contradictions, listing gaps, and listing sources. There is no functional overlap, so an agent can reliably choose the right tool based on the task.

Naming Consistency4/5

The list_* tools are consistent, but get_fragment, draft, and ask_brain deviate from a uniform verb_noun pattern. Despite this, all names are lowercase with underscores and the verbs are descriptive, making the set readable and predictable overall.

Tool Count5/5

With six tools, the server is well-scoped for its purpose of analyzing and synthesizing project sources. Each tool earns its place, and the count is within the ideal 3-15 range.

Completeness4/5

The tool set covers the main workflows: inspecting sources, retrieving fragments, querying historical context, surfacing contradictions and gaps, and drafting. A minor gap is the lack of a tool to list all fragments or ingest new sources, but the server appears intentionally read-only, so this is an acceptable limitation.

Maintenance

ActivitySlowing

ResponsivenessNo issues

Resources

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

AmbixOAuth
ai.ambix
Your product team's shared strategic memory — an MCP server your AI tools reason over.
Knowledge & Memory Product Management
2
PlanExe A
io.github.PlanExeOrg
MCP server for generating rough-draft project plans from natural-language prompts.
Autonomous Agents Developer Tools Project Management
113399
Living MemoryOAuth
to.viibe
An MCP memory server. One memory your agents share — across models, devices and apps.
Knowledge & Memory Autonomous Agents AI & Machine Learning
22
gread A
io.github.NitroRCr
An MCP server that gives your AI access to the source code and docs of all public github repos
Version Control Code Analysis Documentation Access
5361

Related MCP Servers

Dedalus MCP Documentationofficial
Documentation Access Search RAG Systems
dedalus-labs
A
license
A
quality
D
maintenance
An MCP server that serves documentation and enables AI-powered search, Q\&A, and document analysis for developer tools and guides.
Updated a year ago (2025-09-20 12:06 UTC)
5
4
MIT
cortex-brain
Knowledge & Memory Search Documentation Access
sani-savaliya
A
license
Not graded
quality
D
maintenance
An MCP server that enables AI agents to search, read, and contribute to a structured markdown knowledge base with citations, freshness tracking, and a safe write path, providing a shared, auditable company memory.
Updated 3 months ago (2026-06-13 03:16 UTC)
16
MIT
graphmemory
graph-memory
A
license
Not graded
quality
B
maintenance
An MCP server that builds a semantic graph memory from a project directory, indexing documentation and code into graph structures and exposing 70+ MCP tools for search, knowledge management, task management, and more.
Updated 5 months ago (2026-04-09 11:06 UTC)
13
15
Elastic 2.0
Enterprise MCP Documentation
Documentation Access Developer Tools
hasecon
A
license
Not graded
quality
D
maintenance
A self-hosted MCP server that provides up-to-date documentation for enterprise and development tools directly to AI coding assistants like Claude Code and Cursor.
Updated a year ago (2025-07-14 21:49 UTC)
MIT

Quickstart

Tools

What comes back

Two things it will not do

Sample data, and real projects

Development

Links

License

Available Tools

Tool Schema Changelog

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

Dedalus MCP Documentationofficial