# SatRank — Lightning trust + audit oracle (npm · satrank-mcp)

Lightning trust + audit oracle. Score, pay, and audit L402 endpoints with Ed25519 receipts.

- Trust score: 61/100 (medium)
- Change this week: +19
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `satrank-mcp`: 61/100 (this document), [markdown](https://verifymcp.io/servers/dev-satrank-mcp/satrank-mcp.md), [page](https://verifymcp.io/servers/dev-satrank-mcp/satrank-mcp)

## Channel facts

- Registry: `npm`
- Package: `satrank-mcp`
- Version: `1.0.1`
- 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 (106 of 110), 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 (106 of 110), 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 85 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 49/100
  - AI-judged instruction clarity (fair).
  - Tool/resource definitions use about 1227 tokens (~76/item across 16 items; 16 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 69/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 6% 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 dev-satrank-mcp -- npx -y satrank-mcp
```

### Codex

```bash
codex mcp add dev-satrank-mcp -- npx -y satrank-mcp
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add dev-satrank-mcp --command npx --arg -y --arg satrank-mcp
```

### Hermes

```yaml
mcp_servers:
  dev-satrank-mcp:
    command: "npx"
    args: ["-y", "satrank-mcp"]
```

### Other

```json
{
  "mcpServers": {
    "dev-satrank-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "satrank-mcp"
      ]
    }
  }
}
```

## 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 61, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-02 (score 60, +55)

- [security regression] Provenance: unverified → fail
- [security improvement] Install scripts: unverified → pass
- [security improvement] Known CVEs: unverified → partial
- [security improvement] Malware scan: unverified → pass
- [functional improvement] License: unverified → pass
- [functional improvement] Schema quality: unverified → fair
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] Maintenance: unverified → pass
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Stability: unverified → 0.23
- [functional improvement] Tool coverage: unverified → 100
- [functional] Licence: MIT

### 2026-08-01 (score 5, −28)

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

### 2026-07-31 (score 33, −9)

- [security regression] Provenance: unverified → fail
- [security regression] Malware scan: pass → unverified
- [security improvement] Install scripts: unverified → pass
- [functional improvement] License: unverified → pass
- [functional improvement] Maintenance: unverified → pass
- [functional] Licence: MIT

### 2026-07-27 (score 42)

First indexed and scored.

## MCP tools (16)

### `intent` (~167 tokens)

Resolve an intent (category + keywords + budget + latency ceiling) into a ranked list of L402 endpoint candidates from the SatRank catalogue. Returns Bayesian p_success + ci95 + 5-stage breakdown per candidate. Use this BEFORE paying any L402 endpoint to pick the most trusted candidate that fits your budget + SLA. Free to call ; pass `fresh: true` to force a sync HTTP probe (costs 2 sats via /api/intent?fresh=true).

Input parameters:

- `budget_sats` (number)
- `category` (string, required): Service category, e.g. "data/finance", "ai/text", "ai/classify"
- `keywords` (array)
- `limit` (number)
- `max_latency_ms` (number)
- `optimize` (string)

### `get_endpoint_score` (~90 tokens)

Read the public scoring snapshot for a specific L402 endpoint URL. Returns Bayesian p_success + ci95, 5-stage breakdown (challenge / invoice / payment / delivery / quality), median_latency_ms, last_probe_age_sec, freshness_status, source attribution. Use BEFORE calling fulfill on a specific URL to verify trust independently of the ranking.

Input parameters:

- `url_hash` (string, required): sha256 of the endpoint URL

### `fulfill` (~191 tokens)

Pay an L402 endpoint via SatRank fulfill proxy with hold-invoice + auto-issued evidence receipt. Agent supplies intent + max_sats budget + max_latency_ms SLA ; SatRank picks best candidate, settles the L402 payment, returns body + body_sha256 + preimage + operator_pubkey + premium_sats + job_id. Refund on candidate failure. Use this when you want SatRank to write the post-pay evidence trail (Ed25519-signed). NIP-98 auth required ; agent must have a token_balance via /api/deposit OR a bond. The MCP server forwards Authorization headers from the agent runtime.

Input parameters:

- `intent` (object, required)
- `max_latency_ms` (integer, required)
- `max_sats` (integer, required)
- `mode` (string)
- `recall_body` (string)
- `recall_headers` (object)
- `refund_bolt11` (string)

### `fulfill_evidence` (~63 tokens)

Fetch the public Ed25519-signed evidence receipt for a fulfill_jobs row by job_id. Returns canonical_json + signature + body_sha256 + operator_pubkey + ts_started + ts_finished + sats_paid. No auth required.

Input parameters:

- `job_id` (string, required)

### `mini_llm_classify` (~92 tokens)

SatRank-operated L402 endpoint : single-best-label classification (10 sats). Powered by Claude Haiku 4.5 server-side. Use when the public AI catalogue is over-priced (>30 sats/req) for classification. The MCP server forwards Authorization headers ; agent runtime supplies the L402 macaroon after paying the 402 challenge.

Input parameters:

- `labels` (array)
- `text` (string, required)

### `mini_llm_summarize` (~68 tokens)

SatRank-operated L402 endpoint : plain-prose summarization (10 sats). Powered by Claude Haiku 4.5 server-side. Use when the public AI catalogue prices summarization above your budget.

Input parameters:

- `max_words` (integer)
- `text` (string, required)

### `mini_llm_translate` (~49 tokens)

SatRank-operated L402 endpoint : translation (10 sats). POST text + target language code, get back translated text only.

Input parameters:

- `target` (string, required)
- `text` (string, required)

### `aeps.daily_anchor` (~55 tokens)

Fetch the daily Merkle anchor for a UTC day. Returns operator_pubkey + root_hex + receipt_count + L1 broadcast info. Verifiers use this for receipt inclusion proofs.

Input parameters:

- `day_utc` (string, required)

### `aeps.recent_anchors` (~37 tokens)

List the most recent N daily Merkle anchors. Default 30, max 366.

Input parameters:

- `limit` (number)

### `aeps.inclusion_proof` (~88 tokens)

Build the RFC 6962 audit path for a single evidence receipt against the day-Merkle root. Returns root_hex + audit_path[] + leaf_index + tree_size. Verifiers feed these to the AEPS Merkle verifier (TS or Rust impl) plus the L1 anchor txid to confirm the receipt is committed on-chain.

Input parameters:

- `receipt_id` (integer, required)

### `aeps.evidence_receipt` (~47 tokens)

Fetch the Ed25519-signed evidence receipt for a fulfilled job. Same content as fulfill_evidence, addressable by job_id.

Input parameters:

- `job_id` (string, required)

### `aeps.get_dispute` (~49 tokens)

Read the current state of an AEPS dispute. Returns dispute_id + type + multiplier + oracle_pubkeys + oracle_threshold + state + attestations.

Input parameters:

- `dispute_id` (string, required)

### `aeps.list_forks` (~42 tokens)

List detected fork events (operator equivocations). Optional operator_pubkey filter.

Input parameters:

- `limit` (number)
- `operator_pubkey` (string)

### `aeps.get_observations` (~59 tokens)

Return all observed daily anchors for a (operator, day) bucket, grouped by root_hex. distinct_roots > 1 = equivocation.

Input parameters:

- `day_utc` (string, required)
- `operator_pubkey` (string, required)

### `aeps.get_multihop` (~40 tokens)

Read multi-hop HTLC chain state. Returns chain_id + agent_pubkey + per-leg state.

Input parameters:

- `chain_id` (string, required)

### `verify_assertion` (~90 tokens)

Verify offline a SatRank-compatible Nostr trust assertion (kind 30782) or oracle calibration (kind 30783). Validates Schnorr signature + structure + valid_until tag. No network call. Use to compose oracle output across agents without re-querying SatRank.

Input parameters:

- `event` (object, required)
- `expected_oracle_pubkey` (string)
- `now_sec` (number)

## Diagnostics

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

## Score history

- 2026-08-03: 61
- 2026-08-02: 60
- 2026-08-01: 5
- 2026-07-31: 33
- 2026-07-29: 42
- 2026-07-28: 42
- 2026-07-27: 42

## Links

- npm package: https://www.npmjs.com/package/satrank-mcp
- Socket report: https://socket.dev/npm/package/satrank-mcp
- Repository: https://github.com/proofoftrust21/satrank
- Website: https://satrank.dev/
- Changelog RSS feed: https://verifymcp.io/servers/dev-satrank-mcp/satrank-mcp/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/dev-satrank-mcp/satrank-mcp/changelog.json
- HTML version of this page: https://verifymcp.io/servers/dev-satrank-mcp/satrank-mcp