hoop (hoophq/hoop) is an MCP server listed on the M8ven Trust Index. It scores 54 out of 100, grade D. It declares 59 tools. No publisher has claimed this listing.
One gateway in front of every protocol. Same policy across MCP, LLMs, databases and containers. Wire-level enforcement at under 5ms.
Warning. Serious findings were identified. Review the full report before connecting. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
hoophq
Source: github_repo_search
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
serverinfo_getGet gateway server information including version, auth method, license info, and feature flags
access_control_setCanonical way to restrict which user groups can access a connection — matches the webapp 'Access Control' page.
access_control_unsetRemove the access_control binding for a connection. After this call,
access_control_getGet the list of user groups allowed to access a given connection via the access_control plugin
access_control_listList all connections currently bound to the access_control plugin with their allowed user groups
access_request_rules_listList all access request rules for the organization with optional pagination
access_request_rules_getGet a single access request rule by name
access_request_rules_createCreate an approval / just-in-time access workflow rule for one or more connections — matches the webapp 'Access Request' page.
access_request_rules_updateUpdate an existing access request rule by name. Requires admin access
access_request_rules_deleteDelete an access request rule by name. Requires admin access. This action is irreversible
attributes_listList attributes for the organization. Attributes are tags that group connections, access-request/guardrail/data-masking rules, and access-control groups.
attributes_getGet a single attribute by name, including all of its associations (connections, rules, access-control groups).
attributes_createCreate a new attribute, optionally associating it with connections, rules, and access-control groups in the same call. Requires admin access.
attributes_updateUpdate an attribute's description and/or its associations. Requires admin access.
attributes_deleteDelete an attribute by name. Requires admin access. This action is irreversible and removes all of the attribute's associations.
connections_listList all connections accessible to the authenticated user, with optional filters by agent, type, subtype, tag_selector, or search query
connections_getGet a single connection by its name or ID
connections_createCreate a new connection. Requires admin access.
connections_updateUpdate an existing connection by name. Requires admin access. Only provided fields are updated
connections_deleteDelete a connection by name. Requires admin access. This action is irreversible
datamasking_rules_listList all data masking rules configured for the organization
datamasking_rules_getGet a single data masking rule by its ID
datamasking_rules_createCreate a new data masking rule with entity type detection configuration
datamasking_rules_updateUpdate an existing data masking rule by ID
datamasking_rules_deleteDelete a data masking rule by ID. This action is irreversible
execRun a one-shot command or query against a Hoop connection on behalf of the authenticated user.
guardrails_listList all guardrail rules configured for the organization
guardrails_getGet a single guardrail rule by its ID or name
guardrails_createCreate a new guardrail rule with input/output configurations and connection associations
guardrails_updateUpdate an existing guardrail rule by ID
guardrails_deleteDelete a guardrail rule by ID. This action is irreversible
me_getGet the authenticated user's profile: email, name, user_id, org_id,
reviews_listList all reviews (access requests) for the organization
reviews_getGet a single review by its ID or session ID
reviews_updateUpdate a review status (approve, reject, or revoke). Requires membership in a reviewer group
reviews_executeExecute a query that was blocked on an approved one-time review. Runs the originally
reviews_waitLong-poll a review until it reaches a terminal status (APPROVED, REJECTED, REVOKED,
runbook_rules_listList all runbook rules configured for the organization
runbook_rules_getGet a single runbook rule by its ID
runbook_rules_createCreate a new runbook rule with user groups, connections, and allowed runbooks
runbook_rules_updateUpdate an existing runbook rule by ID
runbook_rules_deleteDelete a runbook rule by ID. This action is irreversible
connection_databasesList databases available on a Hoop database connection (Postgres, MySQL, MSSQL, MongoDB).
connection_tablesList tables in a given database on a Hoop database connection.
connection_columnsList columns of a specific table on a Hoop database connection, with types.
sessions_listsessions_getGet session details and metadata by ID. Returns metadata only, not session content or binary data
sessions_get_contentGet the input script and stdout/stderr output of a session.
sessions_get_analysisGet Hoop's AI analysis for a session: risk level, title, explanation, recommended action.
sessions_wait_analysisLong-poll until Hoop's AI analysis for a session is available, or the timeout elapses.
usergroups_listList all user groups in the organization
usergroups_createCreate a user group, optionally binding it to connections and/or existing attributes in the same call.
usergroups_deleteDelete a user group by name. Cannot delete the admin group. This action is irreversible
users_listList all users in the organization
users_getGet a single user by their email address
search_past_sessionsReview the current user's recent sessions to judge whether the behavior under analysis is routine or anomalous.
LowRiskAISessionAnalyzerSelect when the input is safe to execute: non-destructive, scoped, and low operational/security impact.
MediumRiskAISessionAnalyzerSelect when the input could cause performance issues, service disruption, sensitive exposure, or risky-but-not-clearly-destructive changes.
HighRiskAISessionAnalyzerSelect when the input is destructive, irreversible, escalates privileges, exfiltrates data, disables defenses, or resembles exploit/persistence behavior.
API_URLSEGMENT_WRITE_KEYVITE_CLJS_URLOCR_ALLOW_CPU_FALLBACKOCR_REC_LANGOCR_REC_PRECISIONOCR_CUDA_ARENA_STRATEGYOCR_DET_DOWNSCALEGITHUB_OUTPUTTool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
59/59 tools missing one or more hints — serverinfo_get (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); access_control_set (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); access_control_unset (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +56 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Tool test coverage
Only 5/59 tools referenced in tests (8%)
Write tests that reference each tool by name so every tool has at least one test.
No eval / new Function
2 eval() or new Function() calls — dynamic code execution
Replace eval / Function with explicit parsing or safer alternatives.
Readable source code
1 file are minified or bundled, which is usually build output rather than concealment
Ship unminified, readable source.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/hoophq/hoop)?variant=verified from the URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check