# io.github.ozten/cantrip (npm · mcp-server-cantrip)

MCP server for Cantrip — AI-powered GTM engine for solo founders

- Trust score: 68/100 (medium)
- Change this week: +22
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `mcp-server-cantrip`: 68/100 (this document), [markdown](https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip.md), [page](https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip)

## Channel facts

- Registry: `npm`
- Package: `mcp-server-cantrip`
- Version: `1.0.5`
- Transport: `stdio`

## Trust breakdown

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. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Supply Chain Security**: 87/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (98 of 102), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (98 of 102), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 145 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 71/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1758 tokens (~92/item across 19 items; 19 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add ozten-cantrip -- npx -y mcp-server-cantrip
```

### Codex

```bash
codex mcp add ozten-cantrip -- npx -y mcp-server-cantrip
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ozten-cantrip": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "mcp-server-cantrip"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ozten-cantrip --command npx --arg -y --arg mcp-server-cantrip
```

### Hermes

```yaml
mcp_servers:
  ozten-cantrip:
    command: "npx"
    args: ["-y", "mcp-server-cantrip"]
```

### Other

```json
{
  "mcpServers": {
    "ozten-cantrip": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-cantrip"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-02 (score 67, +46)

- [security regression] Provenance: unverified → fail
- [security improvement] Install scripts: unverified → pass
- [security improvement] Known CVEs: unverified → partial
- [security improvement] Malware scan: unverified → pass
- [functional regression] Security disclosure: fail → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] License: unverified → pass
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] Stability: unverified → 0.20
- [functional improvement] Maintenance: unverified → pass
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Schema quality: unverified → good
- [functional] First check of Schema quality: unverified
- [functional] Licence: MIT

### 2026-07-31 (score 21, −7)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 28, −18)

- [security regression] Malware scan: pass → unverified

### 2026-07-27 (score 46)

First indexed and scored.

## MCP tools (19)

### `cantrip_connect` (~66 tokens)

Start here. Connect this workspace to a Cantrip project by writing a .cantrip.json file. All subsequent commands will target this project automatically. Call without arguments to check the current connection.

Input parameters:

- `project` (string): Project name to connect to. Omit to check current connection.

### `cantrip_status` (~61 tokens)

Check daemon health, authentication, and current project. Returns daemon reachability, authenticated identity (user, team), and the active project from .cantrip.json. When CANTRIP_API_KEY is missing, returns setup instructions with config examples for common MCP clients.

### `cantrip_init` (~178 tokens)

Create a new project and connect this workspace to it. Pass 'brief_text' (product brief as text) to auto-extract ICPs, pain points, and value props as inferred entities (costs 5 credits). Or pass 'brief_path' (absolute file path) and the file will be read locally. Without a brief, the project is created empty (free) and you add entities manually. Writes .cantrip.json automatically after creation. After creating a project, add a few entities and confirm them with the user before going deeper.

Input parameters:

- `brief_path` (string): Absolute path to a product brief file — will be read locally and sent as text
- `brief_text` (string): Product brief content as text (preferred)
- `description` (string, required): One-line project description
- `name` (string, required): Project name

### `cantrip_project` (~131 tokens)

Manage projects. Actions:
\- list (default): List all projects in the current team.
\- update: Update a project's name or description. Uses the connected project unless 'slug' is provided.
\- delete: Delete a project and all its data. Uses the connected project unless 'slug' is provided.

Input parameters:

- `action` (string): Action to perform (default: list)
- `description` (string): New description (update only)
- `name` (string): New display name (update only)
- `slug` (string): Project slug to target. Defaults to the connected project from .cantrip.json.

### `cantrip_snapshot` (~132 tokens)

Browse project data at three zoom levels. No args: project overview with entity counts by type and review state, gaps, and review queue size. entity_type only: list all entities of that type (e.g. 'icps', 'pain-points', 'channels'). entity_type + entity_id: show full detail for one entity. This is the primary tool for listing and inspecting entities.

Input parameters:

- `entity_id` (string): Specific entity ID for detail view
- `entity_type` (string): Entity type to drill into (e.g. 'icps', 'pain-points', 'value-props')

### `cantrip_review` (~23 tokens)

List all items pending review (inferred entities and open escalations)

### `cantrip_review_accept` (~33 tokens)

Accept an inferred entity, marking it as verified ground truth

Input parameters:

- `id` (string, required): Entity ID to accept

### `cantrip_review_reject` (~34 tokens)

Reject an inferred entity (soft-delete, kept for history)

Input parameters:

- `id` (string, required): Entity ID to reject

### `cantrip_review_resolve` (~43 tokens)

Resolve an open escalation with a resolution message

Input parameters:

- `id` (string, required): Escalation ID
- `resolution` (string, required): Resolution text explaining the decision

### `cantrip_review_dismiss` (~58 tokens)

Dismiss an inferred entity or open escalation. For entities: removes from review queue without accepting or rejecting (kept for history). For escalations: closes without resolving.

Input parameters:

- `id` (string, required): Entity or escalation ID to dismiss

### `cantrip_next` (~74 tokens)

List gap-analysis opportunities — things that would move the project closer to ideal state. Each opportunity has a stable UUID that you can pass to cantrip_next_prompt or cantrip_next_run. Opportunities persist across calls; re-running gap analysis updates existing opportunities rather than replacing them. Review the project snapshot with the user before running opportunities.

### `cantrip_next_prompt` (~52 tokens)

Generate a context-rich LLM prompt for an opportunity. Returns a ready-to-use prompt with all relevant ontology context baked in. Zero credit cost.

Input parameters:

- `id` (string, required): Opportunity ID from cantrip_next

### `cantrip_next_run` (~104 tokens)

Execute an enrichment opportunity with AI. Runs the LLM-powered enrichment inline — either updating existing entities' missing fields (targeted) or generating new entities (bulk). Returns when complete with a summary of what was created or updated. Parallelism: you may run different loop types concurrently (e.g. enrich ICPs + enrich competitors), but the daemon blocks concurrent runs of the same loop type for safety.

Input parameters:

- `id` (string, required): Opportunity ID from cantrip_next

### `cantrip_history` (~94 tokens)

Query the append-only audit trail of all actions taken on the project

Input parameters:

- `entity` (string): Filter by entity type (e.g. icp, pain_point)
- `limit` (number): Max events to return (default: 50)
- `since` (string): Only events after this ISO date
- `type` (string): Filter by event type (e.g. init, entity_created, review_accept)

### `cantrip_meter_balance` (~31 tokens)

Check remaining credits. Returns available credits, reserved credits (held by in-progress operations), and total balance.

### `cantrip_meter_history` (~42 tokens)

View recent credit transactions. Shows usage debits, purchases, and running balance.

Input parameters:

- `limit` (number): Maximum entries to return (default: 20)

### `cantrip_meter_tiers` (~26 tokens)

View available credit packs. Shows tier name, credits included, and price.

### `cantrip_entity_add` (~363 tokens)

Create a new entity. Automatically marked as 'accepted'. Fields vary by type:
\- icp: name, description, demographics, jobs_to_be_done, willingness_to_pay, current_alternatives, priority, is_beachhead
\- pain_point: description, severity (low|medium|high|critical), frequency (rare|occasional|frequent|constant), evidence
\- value_prop: framing (required — use instead of 'name'; 'description' is stored in extensions), tagline, evidence
\- channel: name, channel_type, lifecycle_stage (exploring|testing|scaling|maintaining|killed), cac, estimated_reach, conversion_rate (note: 'description' maps to 'notes' column)
\- experiment: title (required — use instead of 'name'), hypothesis, description, status (proposed|designed|active|completed|analyzed|abandoned), success_metrics, outcome_notes, value_prop_id, channel_id
\- competitor: name, description, url, positioning, strengths, weaknesses, pricing_model
\- contact: name, email, phone, company, role, source, url, notes
Extra fields (any field not in the schema above) are stored in extensions. After adding entities, pause and confirm with the user before adding more.

Input parameters:

- `description` (string): Entity description
- `entity_type` (string, required): Entity type: icp, pain_point, value_prop, experiment, channel, competitor, contact
- `fields` (object): Additional fields as key-value pairs (e.g. {severity: 'high', frequency: 'constant'})
- `name` (string): Entity name (mapped to 'framing' for value_prop, 'title' for experiment)

### `cantrip_entity_edit` (~138 tokens)

Edit an existing entity. Fields vary by type (same as cantrip_entity_add). Pass well-known fields directly, and any additional fields in the 'fields' object. Extra fields are stored in extensions.

Input parameters:

- `description` (string): Updated description
- `entity_type` (string, required): Entity type: icp, pain_point, value_prop, experiment, channel, competitor, contact
- `fields` (object): Additional fields to update as key-value pairs
- `id` (string, required): Entity ID to edit
- `name` (string): Updated name (mapped to 'framing' for value_prop, 'title' for experiment)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip#diagnostics

## Score history

- 2026-08-03: 68
- 2026-08-02: 67
- 2026-08-01: 21
- 2026-07-31: 21
- 2026-07-30: 28
- 2026-07-28: 46
- 2026-07-27: 46

## Links

- npm package: https://www.npmjs.com/package/mcp-server-cantrip
- Socket report: https://socket.dev/npm/package/mcp-server-cantrip
- Repository: https://github.com/ozten/mcp-server-cantrip
- Changelog RSS feed: https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip/changelog.json
- HTML version of this page: https://verifymcp.io/servers/ozten-cantrip/mcp-server-cantrip