Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

PHION Agent Trust Infrastructure

REMOTE · PHION.SYSTEMS · SCANNED SEP 25

118 agent services with free discovery, intent, x402 terms and signed verification.

Available components

+4 this week 66 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security63
Transport & Reachability100
Schema Quality & AI Usability58
  • AI-judged instruction clarity (fair).Partial
  • Tool/resource definitions use about 5964 tokens (~48/item across 124 items; 124 tools + 0 resources), lean.Pass
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management47
  • Stability observed for 14 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage72
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 2% of tool parameters carry a description.Partial
  • Structured output schemas are declared (58% of tools); any adoption earns full credit.Pass
Tool Safety88
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • 1 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "duplicate_charge_detector" implies "charge" and declares readOnlyHint instead, contradicting what its own name says it does. See how to fix → Partial
  • An AI judge read all 124 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities60
  • Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28. See how to fix → Fail
Install

How do I install the PHION Agent Trust Infrastructure MCP server?

PHION Agent Trust Infrastructure is a hosted endpoint at https://phion.systems/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · phion.systems

# add to Claude Code
claude mcp add --transport http systems-phion-evidence-engine 'https://phion.systems/mcp'
// .cursor/mcp.json
{
  "mcpServers": {
    "systems-phion-evidence-engine": {
      "url": "https://phion.systems/mcp"
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "systems-phion-evidence-engine": {
      "type": "http",
      "url": "https://phion.systems/mcp"
    }
  }
}
# ~/.codex/config.toml
[mcp_servers.systems-phion-evidence-engine]
url = "https://phion.systems/mcp"
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "systems-phion-evidence-engine": {
      "type": "remote",
      "url": "https://phion.systems/mcp",
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add systems-phion-evidence-engine --url 'https://phion.systems/mcp' --transport streamable-http
# ~/.hermes/config.yaml
mcp_servers:
  systems-phion-evidence-engine:
    url: "https://phion.systems/mcp"
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "systems-phion-evidence-engine": {
      "Transport": "http",
      "Url": "https://phion.systems/mcp"
    }
  }
}
# add to Vellum
assistant mcp add systems-phion-evidence-engine -t streamable-http -u 'https://phion.systems/mcp'
// mcp.json
{
  "mcpServers": {
    "systems-phion-evidence-engine": {
      "type": "http",
      "url": "https://phion.systems/mcp"
    }
  }
}

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

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.

  • 25 Sept 26 −1
    • MCP protocol: pass → fail ▼ functional
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
    • Server version: 1.49.4 → 1.49.6 functional
    • Server version: 1.49.3 → 1.49.4 functional
    • Server version: 1.49.0 → 1.49.3 functional
  • 24 Sept 26 +1
    • Server version: 1.46.2 → 1.46.4 functional
  • 23 Sept 26 0
    • Server version: 1.46.1 → 1.46.2 functional
    • Server version: 1.46.0 → 1.46.1 functional
    • Server version: 1.45.0 → 1.46.0 functional
  • 22 Sept 26 +1
    • Schema quality: 4804 → 5964 ▼ functional
    • Tool coverage: 45% → 58% ▲ functional
    • Tool coverage: 45% → 50% ▲ functional
    • Server version: 1.44.0 → 1.45.0 functional
    • Server version: 1.39.0 → 1.44.0 functional
    • Server version: 1.37.0 → 1.39.0 functional
    • New tool “a2a_transaction_bridge” functional
    • New tool “agent_economic_graph” functional
    • New tool “ap2_mandate_bridge” functional
    • New tool “automated_dispute_bundle” functional
    • New tool “autonomous_procurement” functional
    • New tool “cross_protocol_receipt” functional
    • New tool “delegated_spend_policy” functional
    • New tool “dynamic_service_pricing” functional
    • New tool “erc8004_identity_evidence” functional
    • New tool “erc8004_reputation_intelligence” functional
    • New tool “market_demand_predictor” functional
    • New tool “mcp_transaction_gateway” functional
    • New tool “multi_agent_escrow” functional
    • New tool “price_discovery_engine” functional
    • New tool “proof_of_service” functional
    • New tool “reputation_update_receipt” functional
    • New tool “service_gap_detector” functional
    • New tool “sla_risk_predictor” functional
    • New tool “transaction_recovery_v2” functional
    • New tool “agent_behavior_fingerprint” functional
    • New tool “capability_benchmark” functional
    • New tool “capability_verification” functional
    • New tool “economic_loop_detector” functional
    • New tool “payment_optimizer” functional
    • New tool “provider_quality_predictor” functional
    • New tool “sybil_reputation_guard” functional
    • New tool “transaction_risk_score” functional
    • New tool “trust_anomaly_detector” functional
    • New tool “x402_v2_router” functional
  • 21 Sept 26 +2
    • Schema quality: 57 → 50 ▲ functional
    • Tool coverage: 31% → 45% ▲ functional
    • Tool coverage: 31% → 39% ▲ functional
    • MCP protocol: fail → pass ▲ functional
    • MCP protocol version: 2025-06-18 → 2026-07-28 functional
    • Server version: 1.35.0 → 1.37.0 functional
    • Server version: 1.33.0 → 1.35.0 functional
    • Server version: 1.32.2 → 1.33.0 functional
    • Server version: 1.32.3 → 1.32.2 functional
    • New tool “tool_result_schema_validator” functional
    • New tool “agent_memory_provenance” functional
    • New tool “agent_rate_limit_negotiator” functional
    • New tool “agent_session_continuity” functional
    • New tool “cross_agent_receipt_bundle” functional
    • New tool “delegated_credential_guard” functional
    • New tool “execution_cost_estimator” functional
    • New tool “payment_delivery_atomicity” functional
    • New tool “service_failover_selector” functional
    • New tool “task_lease_guard” functional
    • New tool “agent_approval_relay” functional
    • New tool “capability_negotiation_preflight” functional
    • New tool “durable_agent_task” functional
    • New tool “mcp_2026_compatibility_gateway” functional
    • New tool “mcp_a2a_task_bridge” functional
    • New tool “mcp_catalog_cache_guard” functional
    • New tool “oauth_issuer_binding_evidence” functional
    • New tool “quote_freshness_guard” functional
    • New tool “task_cancel_assurance” functional
    • New tool “task_checkpoint_evidence” functional
  • 20 Sept 26 0
    • Server version: 1.32.0 → 1.32.3 functional
  • 19 Sept 26 +1
    • Tool “company_enrichment_evidence” rewrote its description, which is the text the model reads security
    • Tool “contact_enrichment_evidence” rewrote its description, which is the text the model reads security
    • Tool “person_enrichment_evidence” rewrote its description, which is the text the model reads security
    • Schema quality: 50 → 57 ▼ functional
    • Server version: 1.31.1 → 1.32.0 functional
    • “company_enrichment_evidence” added an optional parameter “birth_date” cosmetic
    • “company_enrichment_evidence” added an optional parameter “contact” cosmetic
    • “company_enrichment_evidence” added an optional parameter “country” cosmetic
    • “company_enrichment_evidence” added an optional parameter “email” cosmetic
    • “company_enrichment_evidence” added an optional parameter “email_hash” cosmetic
    • “company_enrichment_evidence” added an optional parameter “entity” cosmetic
    • “company_enrichment_evidence” added an optional parameter “first_name” cosmetic
    • “company_enrichment_evidence” added an optional parameter “identifier” cosmetic
    • “company_enrichment_evidence” added an optional parameter “input” cosmetic
    • “company_enrichment_evidence” added an optional parameter “last_name” cosmetic
    • “company_enrichment_evidence” added an optional parameter “lid” cosmetic
    • “company_enrichment_evidence” added an optional parameter “locality” cosmetic
    • “company_enrichment_evidence” added an optional parameter “minimum_likelihood” cosmetic
    • “company_enrichment_evidence” added an optional parameter “pdl_id” cosmetic
    • “company_enrichment_evidence” added an optional parameter “person” cosmetic
    • “company_enrichment_evidence” added an optional parameter “phone” cosmetic
    • “company_enrichment_evidence” added an optional parameter “postal_code” cosmetic
    • “company_enrichment_evidence” added an optional parameter “query” cosmetic
    • “company_enrichment_evidence” added an optional parameter “region” cosmetic
    • “company_enrichment_evidence” added an optional parameter “school” cosmetic
    • “company_enrichment_evidence” added an optional parameter “street_address” cosmetic
    • “company_enrichment_evidence” added an optional parameter “subject” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “birth_date” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “contact” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “country” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “email_hash” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “entity” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “first_name” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “identifier” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “input” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “last_name” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “lid” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “locality” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “pdl_id” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “person” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “phone” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “postal_code” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “query” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “region” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “school” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “street_address” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “subject” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “ticker” cosmetic
    • “contact_enrichment_evidence” added an optional parameter “website” cosmetic
    • “person_enrichment_evidence” added an optional parameter “birth_date” cosmetic
    • “person_enrichment_evidence” added an optional parameter “contact” cosmetic
    • “person_enrichment_evidence” added an optional parameter “country” cosmetic
    • “person_enrichment_evidence” added an optional parameter “email_hash” cosmetic
    • “person_enrichment_evidence” added an optional parameter “entity” cosmetic
    • “person_enrichment_evidence” added an optional parameter “first_name” cosmetic
    • “person_enrichment_evidence” added an optional parameter “identifier” cosmetic
    • “person_enrichment_evidence” added an optional parameter “input” cosmetic
    • “person_enrichment_evidence” added an optional parameter “last_name” cosmetic
    • “person_enrichment_evidence” added an optional parameter “lid” cosmetic
    • “person_enrichment_evidence” added an optional parameter “locality” cosmetic
    • “person_enrichment_evidence” added an optional parameter “pdl_id” cosmetic
    • “person_enrichment_evidence” added an optional parameter “person” cosmetic
    • “person_enrichment_evidence” added an optional parameter “phone” cosmetic
    • “person_enrichment_evidence” added an optional parameter “postal_code” cosmetic
    • “person_enrichment_evidence” added an optional parameter “query” cosmetic
    • “person_enrichment_evidence” added an optional parameter “region” cosmetic
    • “person_enrichment_evidence” added an optional parameter “school” cosmetic
    • “person_enrichment_evidence” added an optional parameter “street_address” cosmetic
    • “person_enrichment_evidence” added an optional parameter “subject” cosmetic
    • “person_enrichment_evidence” added an optional parameter “ticker” cosmetic
    • “person_enrichment_evidence” added an optional parameter “website” cosmetic
    • “company_enrichment_evidence” reworded the description of “company” cosmetic
    • “contact_enrichment_evidence” reworded the description of “company” cosmetic
    • “person_enrichment_evidence” reworded the description of “company” cosmetic
  • 17 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

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 25 Sept 2026 · Probed https://phion.systems/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=phion.systems CN=YE2,O=Let's Encrypt,C=US 30 Jul 2026 28 Oct 2026 ECDSA 256 ECDSA-SHA384 64d09cefb1b3a722bb9f05fcaf93ae899ed
SANs: phion.systems, www.phion.systems
CN=YE2,O=Let's Encrypt,C=US (CA) CN=Root YE,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 ECDSA 384 ECDSA-SHA384 4df3b15dd6c0784c507cd37b58e6f115
CN=Root YE,O=ISRG,C=US (CA) CN=ISRG Root X2,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 ECDSA-SHA384 872165fc34b6e5fba8add5b3705fb53a
CN=ISRG Root X2,O=Internet Security Research Group,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 SHA256-RSA 6c8f1dc727c7117f7baf853ac980f9cd

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of phion.systems. — Not signed

Zone DS Keys Algorithms Outcome
. trust_anchor 20326, 38696 8, 8 Verified
systems. present 9634 8 Verified
phion.systems. absent Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200
Header Value
strict-transport-security max-age=31536000; includeSubDomains; preload
content-security-policy default-src 'self' https://*.cloudflare.com https://gstatic.com; script-src 'self' 'unsafe-inline' 'unsafe-eval' https://*.jsdelivr.net https://*.cloudflare.com https://*.unpkg.com; style-src 'self' 'unsafe-inline' https://googleapis.com; connect-src 'self' https://*.walletconnect.com https://*.cloudflare-eth.com; img-src 'self' data: https://*.phion.systems https://*.reown.com; font-src 'self' https://gstatic.com;
x-content-type-options nosniff
x-frame-options DENY
referrer-policy strict-origin-when-cross-origin

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://phion.systems/mcp Verified 200
http (plaintext) http://phion.systems/mcp HTTPS enforced 301 https://phion.systems/mcp
MCP tools · 124 exposed · ~5,964 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. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
sla_risk_predictor ~39

Estimate SLA failure probability from bounded success and latency evidence.; 0.004 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

social_source_evidence ~50

Attributable public social-source observations with identity limitations; 0.006 USDC.

NameTypeReqDescription
querystring––
source_urlsarrayyes–
subjectobject––

No output schema declared.

No examples provided.

source_backed_search ~43

Literal search over supplied public sources with citations and hashes; 0.004 USDC.

NameTypeReqDescription
querystringyes–
source_urlsarrayyes–

No output schema declared.

No examples provided.

subscription_spend_guard ~56

Bind one recurring charge to approval, merchant, due time and per-charge/period budgets; 0.003 USDC.

NameTypeReqDescription
policyobjectyes–
subscriptionobjectyes–
usageobjectyes–

No output schema declared.

No examples provided.

sybil_reputation_guard ~41

Combine multiple bounded Sybil signals while abstaining without independent evidence.; 0.004 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

task_cancel_assurance ~21

Task Cancel Assurance; deterministic signed assessment over supplied input.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

task_checkpoint_evidence ~22

Task Checkpoint Evidence; deterministic signed assessment over supplied input.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

task_lease_guard ~21

Task Lease Guard; deterministic signed assessment over supplied input.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

tool_call_policy_guard ~50

Bind a proposed tool call to declared intent and execution policy; 0.002 USDC.

NameTypeReqDescription
callobjectyes–
intentobjectyes–
policyobjectyes–

Structured output declared, but exposes no named fields.

No examples provided.

tool_capability_drift_monitor ~26

Detect tool capability or manifest drift; 0.002 USDC.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

tool_output_firewall ~52

Inspect untrusted MCP/tool output before it enters agent context or triggers an action; 0.002 USDC.

NameTypeReqDescription
output–yes–
policyobject––
sourceobject––

Structured output declared, but exposes no named fields.

No examples provided.

tool_result_schema_validator ~22

Tool Result Schema Validator; deterministic signed assessment over supplied input.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

transaction_assurance ~74

End-to-end settlement, timestamp, delivery-hash and deterministic acceptance assurance; 0.015 USDC.

NameTypeReqDescription
acceptancearray––
deliveryobjectyes–
intentobjectyes–
offerobjectyes–
paymentobjectyes–
transaction_idstringyes–

No output schema declared.

No examples provided.

transaction_recovery ~65

Failure classification and non-repeating recovery plan with validated attempt history; 0.010 USDC.

NameTypeReqDescription
assuranceobjectyes–
attemptsarray––
paymentobjectyes–
policyobjectyes–
transaction_idstringyes–

No output schema declared.

No examples provided.

transaction_recovery_v2 ~38

Recommend bounded recovery and reconcile ambiguous payments before retry.; 0.005 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

transaction_risk_score ~39

Estimate evidence-supported loss probability without selling insurance or assuming liability.; 0.004 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

trust_anomaly_detector ~38

Detect material drift between comparable trust vectors without asserting misconduct.; 0.003 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

try_service ~35

Free representative preview, exact price and upgrade instructions for any PHION paid service; no wallet required.

NameTypeReqDescription
servicestringyes–

No output schema declared.

No examples provided.

verified_news_monitor ~40

Literal query evidence across bounded public news sources; 0.006 USDC.

NameTypeReqDescription
querystring––
source_urlsarrayyes–

No output schema declared.

No examples provided.

verified_web_extract ~51

Bounded public web extraction with source, timestamp and hashes; 0.003 USDC.

NameTypeReqDescription
expected_fieldsarray––
querystring––
source_urlsarrayyes–

No output schema declared.

No examples provided.

verify ~35

Free verification of a PHION signed receipt.

NameTypeReqDescription
content–––
receiptobjectyes–
sourcesarray––

No output schema declared.

No examples provided.

webhook_verifier ~63

Fail-closed RFC 9421-style webhook preflight requiring trusted key, algorithm, nonce, freshness, replay status and signature coverage of method, target and content; 0.003 USDC.

NameTypeReqDescription
eventobjectyes–
verificationobjectyes–

No output schema declared.

No examples provided.

x402_quote_comparator ~40

Validate x402 v2 fields and compare only quotes sharing asset and decimals; 0.002 USDC.

NameTypeReqDescription
quotesarrayyes–

No output schema declared.

No examples provided.

x402_v2_router ~43

Select an exactly compatible x402 v2 payment requirement without receiving private keys.; 0.003 USDC. Deterministic, evidence-bounded and signed.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

Common questions

What is the PHION Agent Trust Infrastructure MCP server?

PHION Agent Trust Infrastructure is an MCP server listed in the public MCP registry as systems.phion/evidence-engine. 118 agent services with free discovery, intent, x402 terms and signed verification. This page covers its hosted endpoint (https://phion.systems/mcp).

Is the PHION Agent Trust Infrastructure MCP server safe to use?

PHION Agent Trust Infrastructure scores 66 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the PHION Agent Trust Infrastructure MCP server expose?

PHION Agent Trust Infrastructure exposes 124 tools: phion_execute, index_feed, preflight, try_service, schema_normalize_free, and 119 more. Their descriptions and schemas cost roughly 5,964 tokens of context every time the server is loaded.

Does the PHION Agent Trust Infrastructure MCP server require authentication?

No. We connected to PHION Agent Trust Infrastructure without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the PHION Agent Trust Infrastructure MCP server still maintained?

PHION Agent Trust Infrastructure is still listed as active in the MCP registry. We last reached this channel on 25 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.