asksteps

co.asksteps/asksteps

by co.asksteps· v1.0.0Official MCP Registry

Findings

Detected by Squatch rules from the server's metadata and protocol behavior. Latest completed scan: Nov 13, 2026, 01:41 UTC.

Severity: High3 findings

HighOpenSQ-SERVER-AUTH-NONE-WRITE

State-changing tools available without authentication

The remote server lists state-changing tools without requiring authentication.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Require MCP authorization for servers that expose state-changing tools.

HighOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

HighOpenSQ-TOOL-TITLE-EFFECT-MISMATCH

Display title describes a lower-impact action than the name

The tool's display title describes a lower-impact action than its name (for example a read title on a delete tool).

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Make the title describe the same action as the tool name.

Severity: Medium9 findings

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CREDENTIAL-ARGUMENT

Asks the AI model to supply credentials as arguments

A tool parameter asks the AI model to supply a password, token, key or similar credential, which the model can see and repeat.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Obtain credentials through server-side configuration or MCP authorization instead of tool arguments.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

MediumOpenSQ-TOOL-CONCEALMENT

Text asks the AI model to hide actions from the user

Model-facing text instructs the AI model to hide actions or information from the user.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Remove any wording that asks the model to act silently or withhold information from the user.

Severity: Low9 findings

LowOpenSQ-REPO-MISSING

No source repository linked

No source repository is linked in the registry entry.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Add a repository to server.json.

LowOpenSQ-TOOL-META-INSUFFICIENT

Description too thin to determine behavior

The tool's name and description do not state what it does.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Describe the action the tool performs, its target and its side effects.

LowOpenSQ-CONFORMANCE-MODERN-FAILURE

Protocol conformance checks failed

The server failed one or more MCP protocol conformance checks.

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Review the failed checks against the MCP specification.

LowOpenSQ-TOOL-ANNOTATION-UNDERCLAIM

Annotations understate what the tool does

The tool declares itself read-only, but its name or description indicates it changes, deletes or runs something. Clients that auto-approve read-only tools would run it without confirmation.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Set readOnlyHint to false and destructiveHint to reflect the tool's behavior.

LowOpenSQ-TOOL-ANNOTATION-UNDERCLAIM

Annotations understate what the tool does

The tool declares itself read-only, but its name or description indicates it changes, deletes or runs something. Clients that auto-approve read-only tools would run it without confirmation.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Set readOnlyHint to false and destructiveHint to reflect the tool's behavior.

LowOpenSQ-TOOL-ANNOTATION-UNDERCLAIM

Annotations understate what the tool does

The tool declares itself read-only, but its name or description indicates it changes, deletes or runs something. Clients that auto-approve read-only tools would run it without confirmation.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Set readOnlyHint to false and destructiveHint to reflect the tool's behavior.

LowOpenSQ-TOOL-INJECTION-INSTRUCTIONS

Instructions addressed to the AI model

Model-facing text contains instructions addressed to the AI model rather than a description of the tool (for example to override prior instructions, to act before using tools, or to react to trigger phrases).

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Describe what the tool does and its parameters. Do not include instructions to the model about other tools, the user or the conversation.

LowOpenSQ-TOOL-INJECTION-INSTRUCTIONS

Instructions addressed to the AI model

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Describe what the tool does and its parameters. Do not include instructions to the model about other tools, the user or the conversation.

LowOpenSQ-TOOL-INJECTION-INSTRUCTIONS

Instructions addressed to the AI model

First seen Oct 15, 2026 · Last seen Nov 13, 2026

Suggested fix: Describe what the tool does and its parameters. Do not include instructions to the model about other tools, the user or the conversation.

Severity: Info1 finding

InfoOpenSQ-TOOL-EXFIL-SHAPED-PARAM

Free-text parameter commonly used to collect conversation data

A free-text parameter has a name commonly abused to collect conversation, prompt or environment data.

First seen Nov 13, 2026 · Last seen Nov 13, 2026

Suggested fix: Constrain the parameter (enum, maxLength) or rename it to reflect its purpose; never ask the model to fill it with conversation content.

What this scan does not cover

Squatch never calls a server's tools, so it cannot see what they actually do at run time.
It scans without credentials: tools behind authentication may not be visible.
It does not review source code, dependencies or the publisher's infrastructure.
A server can change behavior between scans; monitoring compares each scan with the last.
Classifications come from tool names, descriptions and schemas, which the publisher controls.

How Squatch scans servers

Security scan

High-severity findings detected

3 open findings at HIGH or above (0 critical).

1 piece of evidence

Source: Squatch observedscan
Metadata and configuration analysis; tools are not executed.
Observed Nov 13, 2026, 01:41 UTC

Last checked Nov 13, 2026, 01:41 UTC

Authentication

Squatch observed

Summary: { "type": "none", "challengeStatus": null, "scopesSupported": [], "resourceMetadataUrl": null, "authorizationServers": [] }

Protocol conformance

Checks: { "checks": [ { "check": "tools_list_deterministic", "passed": true }, { "check": "list_methods_succeed", "passed": true }, { "check": "list_results_schema_valid", "passed": true }, { "check": "pagination_terminates", "passed": true }, { "check": "tool_names_valid", "passed": true }, { "check": "tool_names_unique", "passed": true }, { "check": "no_network_schema_refs", "passed": true }, { "check": "x_mcp_header_valid", "passed": true }, { "check": "icon_urls_https_or_data", "detail": "no icons declared", "passed": true }, { "check": "discover_present", "passed": true }, { "check": "result_type_present", "passed": true }, { "check": "cacheable_fields_present", "passed": true }, { "check": "server_info_in_meta", "passed": true }, { "check": "no_session_id", "passed": true }, { "check": "unknown_method_404", "passed": true }, { "check": "notification_202", "passed": true }, { "check": "get_405", "detail": "probe failed: content_type", "passed": false } ], "toolsListNondeterministic": false }

Full technical details