Robin Saige we check the tools AI agents call

EchoRelay

dev.echorelay/management

Manage EchoRelay API integrations from any MCP client: lines, endpoints, keys, billing, logs, DLQ.

— the operator's own registry description reported

Publisher dev.echorelay · first seen 2026-08-12 · endpoint https://mcp.echorelay.dev

allow
verdict
rich-slow-solo
cluster
ok_tools
answers
78
tools served

Dependency rating derived

allow — rich-slow-solo — safe to depend on

Cluster rich-slow-solo · distinctiveness 0.531 (moderate). Every figure is a percentile or class within the 11,094-server censussee the whole spectrum.

Reach — can an agent get to it?
answersalive_tools
latency416 ms
vs populationslow
protocolcurrent
authopen
Use — can it use it?
tools78
vs populationhigh
capabilities1
harnessA
Trust — can it rely on it?
identitysolo
servers on its host1
verifiabilityunassessed
duplicate inventoryno
drift events0
answers trueno primary

🔏 Signed receipt rr_6b8a6ca12baf42a3481ee228 · ed25519 · key rs-rcpt-2026-08 — this rating is tamper-evident; an agent gets the full signature from check_server and can verify it against the published key.

What kind of tools these are derived

Verification only means something relative to a tool's NATURE — an email-sender has no "true answer", a generator has no answer key. This server's 78 tools, classified, with what checking applies to each kind (the full methodology →):

naturetoolsmeaning what checking applies
action39changes the worldno true answer exists; we check the CONTRACT (destructive/read-only/idempotent declarations, error legibility) and never fire real actions
retrieval-public29serves public factstruth-checkable against a public source — the V-ladder applies in full
unclassified6not yet classifiable from its textTier-1 only until its tools describe themselves
generative2creates contentno answer key exists; disclosure + stability checks, never truth verdicts
computational1deterministic transformsself-checkable by re-computation and known-answer tests
retrieval-private1serves the caller's own dataoutside truth-checking is impossible BY NATURE, not by failure — the operator-invited pathway applies

Tools last seen

activate_line archive_project cancel_line_publish cancel_subscription change_plan config_diff create_api_key create_credential create_endpoint create_line create_project create_project_token deactivate_line default_endpoint_template delete_credential delete_endpoint delete_line discard_dlq_entry discard_draft_endpoint discard_line_draft downgrade_addon dry_run_endpoint duplicate_line get_billing get_config get_dlq_entry get_endpoint get_key_policy get_line_draft get_metrics get_project get_receipts get_request get_subscription invite_member list_active_addons list_addons list_api_keys list_audit_events list_config_revisions list_credentials list_credit_packs list_dlq list_endpoints list_lines list_members list_project_tokens list_projects list_requests list_subscription_plans preview_line_draft preview_plan_change protect_project publish_line_draft remove_member rename_project resend_invite resume_addon retry_dlq_entry reveal_rotation_successor revoke_api_key revoke_project_token rollback_config rotate_api_key schedule_line_publish set_burst_opt_in set_docs_shared set_key_policy set_member_role set_outbound_allowlist set_rpm_ceiling start_subscription start_topup subscribe_addon unarchive_project unsubscribe_addon update_credential update_endpoint

Harness readiness observed

Can an agent's harness pick this tool and call it correctly? Graded on the three things an agent reads — name, description, typed parameters. Band A — an agent can pick and call these reliably.

legibility checktools passingok
name is clear & specific78/78
has a description78/78
description is substantive78/78
parameters are typed78/78
parameters are described68/7868/78

78 of 78 tools graded · 78 with a captured input schema. RS-008, Tier-1 — no domain knowledge, applies to any tool. “Alive” is not the same as “usable”.

Truth checks observed

Not in the Tier-2 trade lane, or not yet checked. Tier-2 re-derives a server's answers against published primary sources — see truth checks.

Protocol conformance observed

2025-06-18
protocol version
session model
1.13.0
server version echorelay-mcp

Capabilities: tools

Drift

No confirmed drift on record. Baseline set 2026-08-14; changes appear here after human review.

History observed

Every probe we made, oldest → newest (2 shown, 2026-08-12 → 2026-08-14). Green answered · amber answered-but-walled · red no useful answer. A gap in our cadence is a gap in coverage, not evidence about the server.

changelog (2 events)
datetypewhat changed
2026-08-14inventorytools 77 → 78
2026-08-12outcomefirst probe: ok_tools
raw probe records
probed (UTC) outcomehttpmsprotocol
2026-08-14T19:58:43Zok_tools2004162025-06-18
2026-08-12T05:33:57Zok_tools2004862025-06-18

Watch or embed this verdict

verdict badge

Operators: put the live verdict in your README — it updates with every census, links back here, and is a dated observation, never a warranty:

[![Robin Saige verdict](https://robinsaige.com/badge/dev.echorelay/management.svg)](https://robinsaige.com/s/dev.echorelay/management)

Depend on it? Subscribe to its change feed — outcome changes and confirmed drift, no account needed. Building on it? The full dossier as JSON — verdict, rating, truth checks, history — stable enough to gate CI on.

← overview