io.github.Savirinc/unfragile

NPM · @UNFRAGILE/MCP-SERVER · SCANNED AUG 3

The match graph for AI. Search 100K+ capabilities across 13K+ AI artifacts.

Available components

+24 this week 64 64 Trust /100

Trust breakdown (6 categories)

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 →

Supply Chain Security87

No malware found by supply-chain analysis.Pass
Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree.Partial
No install/post-install scripts declared.Pass
Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree. View diagnostics → Partial

Provenance & Transparency19

Repository check failed: the declared repository URL returned HTTP 404. See how to fix → View diagnostics → Fail
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 16 days ago).Pass
Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified

Schema Quality & AI Usability77

AI-judged instruction clarity (excellent).Pass
Tool/resource definitions use about 978 tokens (~97/item across 10 items; 10 tools + 0 resources), lean.Pass
Usage-examples check failed: none of the tools include examples. See how to fix → Fail

Stability & Change Management27

Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial

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

Capabilities100

Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass

Install

Add this component to your MCP client. Where a client-specific snippet is available, pick your client below and copy it straight into your config; otherwise use the connection detail shown.

npm · @unfragile/mcp-server

Claude Codex opencode OpenClaw Hermes Other

# add to Claude Code

claude mcp add savirinc-unfragile -- npx -y @unfragile/mcp-server

# add to Codex CLI

codex mcp add savirinc-unfragile -- npx -y @unfragile/mcp-server

// opencode.json

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

# add to OpenClaw

openclaw mcp add savirinc-unfragile --command npx --arg -y --arg @unfragile/mcp-server

# ~/.hermes/config.yaml

mcp_servers:
  savirinc-unfragile:
    command: "npx"
    args: ["-y", "@unfragile/mcp-server"]

// mcp.json

{
  "mcpServers": {
    "savirinc-unfragile": {
      "command": "npx",
      "args": [
        "-y",
        "@unfragile/mcp-server"
      ]
    }
  }
}

Changelog

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.

3 Aug 26 +4
- Stability: unverified → 0.27 ▲ functional
2 Aug 26 +60
- Provenance: unverified → fail ▼ security
- Known CVEs: unverified → partial ▲ security
- Install scripts: unverified → pass ▲ security
- Malware scan: unverified → pass ▲ security
- Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window). security
- Schema quality: unverified → excellent ▲ functional
- Tool coverage: unverified → 100 ▲ functional
- License: unverified → pass ▲ functional
- Dependency health: unverified → partial ▲ functional
- Maintenance: unverified → pass ▲ functional
- MCP protocol: unverified → pass ▲ functional
- Licence: MIT functional
1 Aug 26 −16
- Tool coverage: 100 → unverified ▼ functional
31 Jul 26 +16
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
30 Jul 26 −40
- Malware scan: pass → unverified ▼ security
- Tool coverage: 100 → unverified ▼ functional
- First check of Schema quality: unverified functional
27 Jul 26 40

First indexed and scored.

Show 3 earlier entries Hide earlier entries

Diagnostics

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 3 Aug 2026 · Analysed npm/@unfragile/[email protected]

Provenance none

Ecosystem: npm · Outcome: none

Dependencies 95 packages

95 packages in the resolved dependency tree · 95 deprecated · 29 stale.

The dependency tree was only partially resolved, so these counts may be incomplete.

MCP tools — 10 exposed · ~978 tokens

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.

Tool Tokens

compare ~81

Compare two AI artifacts side-by-side. Shows capabilities, pricing, rank, and graph signals for each. Uses search-based lookup (best-effort name matching). Use this when deciding between alternatives.

Input Output Examples

Name	Type	Req	Description
artifact_a	string	yes	First artifact name (e.g., 'cursor')
artifact_b	string	yes	Second artifact name (e.g., 'windsurf')

No output schema declared.

No examples provided.

feedback ~101

Report whether a recommended tool worked or not. This closes the learning loop — the Unfragile graph uses this feedback to improve future recommendations. Call this after trying a tool from search results.

Input Output Examples

Name	Type	Req	Description
comment	string	—	Optional: brief note on why it worked or didn't
matchRecordId	string	yes	Match record ID from search results (shown at the bottom of search output)
outcome	string	yes	Did the recommended tool work for your use case?

No output schema declared.

No examples provided.

find_mcps ~86

Find MCP servers by capability need. Use this when you need to discover MCP servers for specific integrations (e.g., databases, APIs, cloud services). Returns MCP servers ranked by capability match.

Input Output Examples

Name	Type	Req	Description
limit	number	—	Max results
need	string	yes	What capability you need (e.g., 'Postgres database access', 'Slack messaging', 'GitHub repository management')

No output schema declared.

No examples provided.

find_stack ~149

Assemble a complete AI harness stack for a use case. Given a description of what you're building, returns recommended tools across harness layers: orchestration, tools/MCPs, memory, guardrails, context assembly, and evaluation. This is the key differentiator — Unfragile understands that modern AI systems are composed of 5-15 tools working together.

Input Output Examples

Name	Type	Req	Description
description	string	yes	What you're building (e.g., 'a customer support agent that connects to our Postgres database and Slack, with memory of past conversations')
focus	string	—	Stack focus: 'full' = all layers, 'tools-only' = just MCPs and integrations, 'infrastructure' = frameworks and platforms

No output schema declared.

No examples provided.

get_artifact ~89

Get full details and capabilities for a specific AI artifact by name or slug. Uses search-based lookup (best-effort name matching — may return a different artifact for ambiguous names like 'express'). Use this to understand what an artifact can do before adding it to your stack.

Input Output Examples

Name	Type	Req	Description
name	string	yes	Artifact name or slug (e.g., 'cursor', 'langchain', 'claude-code')

No output schema declared.

No examples provided.

resolve_capability ~126

Resolve a capability:// URI or natural-language capability into ranked AI artifacts. This is Unfragile's capability DNS primitive. Use when an agent knows what capability it needs and must choose the safest current artifact at runtime.

Input Output Examples

Name	Type	Req	Description
capability	string	yes	Capability URI or phrase, e.g. 'capability://database.postgres.query.readonly' or 'read-only Postgres database access'
limit	number	—	Max resolutions
risk	string	—	Risk posture. production/enterprise/strict require stronger quality/status gates.
type	string	—	Filter by artifact type

No output schema declared.

No examples provided.

search ~98

Search the Unfragile match graph for AI tools, frameworks, APIs, MCP servers, agents, and more. Returns ranked results with capability matches and graph signals. Every query feeds the graph.

Input Output Examples

Name	Type	Req	Description
limit	number	—	Max results to return
query	string	yes	What you're looking for (e.g., 'best framework for building AI agents', 'MCP server for database access')
type	string	—	Filter by artifact type

No output schema declared.

No examples provided.

subscribe ~117

Set up a persistent watch for new AI tools matching a query. Get notified daily when something new appears in the Unfragile graph. Requires at least one notification channel (email or webhook).

Input Output Examples

Name	Type	Req	Description
email	string	—	Email address for notifications
query	string	yes	What to watch for (e.g., 'new MCP server for Postgres', 'framework for building AI agents')
type	string	—	Only watch for this artifact type
webhook	string	—	Webhook URL for notifications (receives POST with new matches)

No output schema declared.

No examples provided.

trust_passport ~90

Get the machine-readable trust passport for an AI artifact. Use this before an agent selects a tool, API, MCP server, model, repo, or framework for a task. Returns capability URIs, permissions, data access risk, known failure modes, observed outcomes, and trust score.

Input Output Examples

Name	Type	Req	Description
slug	string	yes	Artifact slug (e.g., 'cursor', 'claude-code', 'langchain')

No output schema declared.

No examples provided.

unsubscribe ~41

Cancel a persistent watch (monitor). Use the monitor ID returned from subscribe.

Input Output Examples

Name	Type	Req	Description
monitorId	string	yes	Monitor ID from subscribe (e.g., 'mon_...')

No output schema declared.

No examples provided.

On this page

Handy links

Scoring key

80+ high trust
60–79 medium
below 60 low

How we score

Score history · last 30 days

27 Jul 3 Aug