Loading…
Designs, prices, and deploys AWS and GCP cloud infrastructure from plain-English requirements.
Do not connect
A critical issue was found. Do not connect this server as-is.
Scanned 1 month ago Due for re-check
A server can change after it's graded. Re-run the automated scan to refresh this report.
This grade is deterministic and reproducible: the same server surface always yields the same grade under a given algorithm version. It is a real automated assessment computed by the MCPGrade engine from what the probe actually observed — not a fabricated or opinion score. It is not a manual human pentest, so it can miss context-specific risks.
Every signal below was measured directly by the automated probe. The grade is derived only from evidence like this — nothing is assumed.
The insideout-agent MCP server exposes 24 tools, focused primarily on general-purpose capabilities. Its published description reads: "Designs, prices, and deploys AWS and GCP cloud infrastructure from plain-English requirements". It communicates over Streamable HTTP using the 2025-06-18 protocol revision, and does not require authorization to connect. MCPGrade currently rates insideout-agent F — a critical issue was found and the server should not be connected as-is. Its most notable findings include "Hidden instructions in a tool description" and "Hidden instructions in a tool description". This report is a deterministic, reproducible automated assessment: the same observed surface always yields the same grade under a given algorithm version, and it is refreshed as new scans arrive and free to read — but it is not a substitute for a manual human security review. Always re-check insideout-agent's advertised tools, transport security, and authorization posture before connecting an autonomous agent or sharing sensitive context with it.
Add insideout-agent to an MCP client using the endpoint below. Review the grade and findings above before granting an autonomous agent access.
MCP endpoint
https://app.luthersystems.com/v1/insideout-mcpStreamable HTTP transport. This is the MCP endpoint, not a website — paste it into your MCP client server list rather than a browser.
Captured passively during the read-only scan. Click any value to find servers that match it.
The tools this server advertises via tools/list — names, purposes, and the parameters each accepts, exactly as enumerated read-only.
INSPECTION: Inspect AWS infrastructure for a deployed project ⚠️ **PREREQUISITE**: This tool requires a prior deployment ATTEMPT (successful or failed). Check convostatus for hasDeployAttempt=true bef
BATCH INSPECTION: run up to 32 AWS inspect probes in one call. ⚠️ **PREREQUISITE**: Same as awsinspect — deploy attempt required. Check convostatus for hasDeployAttempt=true before calling. Use this w
Sorted worst-first. Each finding shows its severity, what it means, its OWASP MCP Top-10 mapping, and a recommended fix — the check id links to the exact methodology row that produced it.
tool: convoopen
A tool description contains imperative text aimed at the model (e.g. "ignore previous", "before answering read …").
rchitect agent. IMPORTANT: source (string) - You MUST set this to identifyRecommendationRemove model-directed instructions from tool descriptions.
tool: credawait
A tool description contains imperative text aimed at the model (e.g. "ignore previous", "before answering read …").
ploy to deploy IMPORTANT: Do NOT call this tool without first showing theRecommendationRemove model-directed instructions from tool descriptions.
tool: tfdeploy
A tool description contains imperative text aimed at the model (e.g. "ignore previous", "before answering read …").
been generated. IMPORTANT: This starts a long-running job (15+ minutes). URecommendationRemove model-directed instructions from tool descriptions.
tool: tfdestroy
A tool description contains imperative text aimed at the model (e.g. "ignore previous", "before answering read …").
ful deployment. IMPORTANT: This starts a long-running job. Use tfstatus/tfRecommendationRemove model-directed instructions from tool descriptions.
tool: awsinspect
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). Supported serviRecommendationDescriptions must describe only their own tool.
tool: awsinspect_batch
A tool description tries to alter the model’s use of another tool.
SITE**: Same as awsinspect — deploy attempt required. Check convostatus foRecommendationDescriptions must describe only their own tool.
tool: convoawait
A tool description tries to alter the model’s use of another tool.
ession_id: from convoopen response OPTIONAL: - message_id: if known (froRecommendationDescriptions must describe only their own tool.
tool: convoinspect
A tool description tries to alter the model’s use of another tool.
e it with: • `convostatus` for the live stack / config / pricing • `tfrRecommendationDescriptions must describe only their own tool.
tool: convoopen
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
tructure design conversation Open an InsideOut V2 session and receive the asRecommendationRemove side-channel parameters; constrain tool inputs.
tool: convoopen
A tool description tries to alter the model’s use of another tool.
ent tool calls (convoreply, tfgenerate, tfdeploy, etc.). ⚠️ The session_idRecommendationDescriptions must describe only their own tool.
tool: convoreply
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
tructure design conversation Send a user message to the active InsideOut sesRecommendationRemove side-channel parameters; constrain tool inputs.
tool: convoreply
A tool description tries to alter the model’s use of another tool.
RecommendationDescriptions must describe only their own tool.
tool: convostatus
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
including the ?token=... suffix (format: sess_v2_*?token=*). The sufRecommendationRemove side-channel parameters; constrain tool inputs.
tool: convostatus
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...).RecommendationDescriptions must describe only their own tool.
tool: credawait
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: cloudRecommendationDescriptions must describe only their own tool.
tool: credawait
The text tells the model WHEN to call this tool relative to others ("always call first", "before any other tool", "chain to X tool") — a toxic-flow injection that hijacks the agent’s orchestration rather than describing the tool.
PORTANT: Do NOT call this tool without first showing the connect URL to the user. The user nRecommendationTool metadata must describe only the tool, never sequence the agent’s calls.
tool: gcpinspect
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). Supported serviRecommendationDescriptions must describe only their own tool.
tool: gcpinspect_batch
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). Supported serviRecommendationDescriptions must describe only their own tool.
tool: help
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
overview (~500 tokens).","enum":["workflow","tools","examples","inspeRecommendationRemove side-channel parameters; constrain tool inputs.
tool: stackdiff
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...).RecommendationDescriptions must describe only their own tool.
tool: stackrollback
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...), version (targetRecommendationDescriptions must describe only their own tool.
tool: stackversions
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...).RecommendationDescriptions must describe only their own tool.
tool: submit_feedback
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
including the ?token=... suffix (format: sess_v2_*?token=*). The sufRecommendationRemove side-channel parameters; constrain tool inputs.
tool: tfdeploy
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
rovably wedged. CREDENTIAL FLOW (if credentials are missing): 1. ResponseRecommendationRemove side-channel parameters; constrain tool inputs.
tool: tfdeploy
A tool description tries to alter the model’s use of another tool.
ying, or pass force_new=true to override. Returns confirmation that the deRecommendationDescriptions must describe only their own tool.
tool: tfdestroy
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
including the ?token=... suffix (format: sess_v2_*?token=*). The sufRecommendationRemove side-channel parameters; constrain tool inputs.
tool: tfdestroy
A tool description tries to alter the model’s use of another tool.
S: session_id from convoopen response (format: sess_v2_...). OPTIONAL: fRecommendationDescriptions must describe only their own tool.
tool: tfdrift
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
including the ?token=... suffix (format: sess_v2_*?token=*). The sufRecommendationRemove side-channel parameters; constrain tool inputs.
tool: tfdrift
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). PREREQUISITE:RecommendationDescriptions must describe only their own tool.
tool: tfgenerate
Tools that execute commands / write files / make arbitrary requests are exposed without annotations or guards.
code inline if curl/unzip fails. **CRITICAL WORKFLOW** (default moRecommendationAnnotate destructive tools and require human approval.
tool: tfgenerate
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: incluRecommendationDescriptions must describe only their own tool.
tool: tflogs
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: job_iRecommendationDescriptions must describe only their own tool.
tool: tfoutputs
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: job_iRecommendationDescriptions must describe only their own tool.
tool: tfplan
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
rovably wedged. CREDENTIAL HANDLING: Same as tfdeploy - credentials must bRecommendationRemove side-channel parameters; constrain tool inputs.
tool: tfplan
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: sanRecommendationDescriptions must describe only their own tool.
tool: tfruns
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...).RecommendationDescriptions must describe only their own tool.
tool: tfstatus
A tool description tries to alter the model’s use of another tool.
session_id from convoopen response (format: sess_v2_...). OPTIONAL: job_iRecommendationDescriptions must describe only their own tool.
The server accepts tool enumeration (and likely invocation) with no authentication.
RecommendationRequire OAuth 2.1 authorization for any server exposing non-public tools.
tool: awsinspect
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: awsinspect_batch
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: convoawait
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: convoinspect
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: convoopen
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: convoreply
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: convostatus
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: credawait
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: gcpinspect
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: gcpinspect_batch
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: help
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: stackdiff
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: stackrollback
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: stackversions
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: submit_feedback
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfdeploy
The server advertises open-world / broadly-scoped capabilities.
annotations.destructiveHint = trueRecommendationScope tools to the minimum needed.
tool: tfdestroy
The server advertises open-world / broadly-scoped capabilities.
annotations.destructiveHint = trueRecommendationScope tools to the minimum needed.
tool: tfdrift
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfgenerate
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfgenerate
One tool both accesses the filesystem and reaches the network. Combining two capability classes in a single tool widens its blast radius and is the substrate for confused-deputy and exfiltration abuse.
you like me to save the Terraform files? Default: ./insideout-infra/' 3. WAIT for userRecommendationSeparate filesystem and network capabilities into distinct, independently-scoped tools.
tool: tflogs
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfoutputs
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfplan
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfruns
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: tfstatus
The server advertises open-world / broadly-scoped capabilities.
annotations.openWorldHint = trueRecommendationScope tools to the minimum needed.
tool: awsinspect
An unusually long description is a common injection-padding tactic.
description length 2287 charsRecommendationKeep descriptions concise.
tool: awsinspect_batch
An unusually long description is a common injection-padding tactic.
description length 2130 charsRecommendationKeep descriptions concise.
tool: convoopen
An unusually long description is a common injection-padding tactic.
description length 2282 charsRecommendationKeep descriptions concise.
tool: convoreply
An unusually long description is a common injection-padding tactic.
description length 2032 charsRecommendationKeep descriptions concise.
tool: gcpinspect
An unusually long description is a common injection-padding tactic.
description length 2615 charsRecommendationKeep descriptions concise.
tool: gcpinspect_batch
An unusually long description is a common injection-padding tactic.
description length 2098 charsRecommendationKeep descriptions concise.
Vantaj uptime monitoring via MCP — manage monitors, heartbeats, incidents, and status pages.
Unified gateway to Algeria's TKAWEN ecosystem: commerce, certification, and AI tools.
Provides access to the Cohereon Doctrine AI safety framework with governance components, tiered access, and agent onboarding.
Agentic rails for complex workflows with receipts, fees, and MCP tool access.
Structural TC39 spec lookup for ECMA-262 and ECMA-402 in AI agents, SHA-pinned and offline-first.
Structural TC39 spec lookup for ECMA-262 and ECMA-402 in AI agents, SHA-pinned and offline-first.
Wait for a pending response from Riley after a convoreply timeout. 🎯 USE THIS TOOL WHEN: convoreply returned a timeout error. This allows you to continue waiting for the response without resending th
INSPECTION: View a session's conversation transcript and metadata Returns the full message history (user / assistant / tool turns) plus the session's meta — workflow step, cloud, deployment status, dr
WORKFLOW: Step 1 of 4 - Start infrastructure design conversation Open an InsideOut V2 session and receive the assistant's intro message. The response contains a clean message from Riley (the infrastru
WORKFLOW: Step 2 of 4 - Continue infrastructure design conversation Send a user message to the active InsideOut session and receive the assistant reply. The response contains a clean message from Rile
INSPECTION: View the current infrastructure stack for a session Returns the current state of the user's infrastructure design including: **Components** - Selected infrastructure services (VPC, databas
Wait for the user to securely connect their cloud account and subscribe to Luther Systems. Polls until credentials appear on the session. 🎯 USE THIS TOOL WHEN: tfdeploy returns an 'auth_required', 'n
INSPECTION: Inspect GCP infrastructure for a deployed project ⚠️ **PREREQUISITE**: This tool requires a prior deployment ATTEMPT (successful or failed). Check convostatus for hasDeployAttempt=true bef
BATCH INSPECTION: run up to 32 GCP inspect probes in one call. ⚠️ **PREREQUISITE**: Same as gcpinspect — deploy attempt required. Check convostatus for hasDeployAttempt=true before calling. Use this w
Get workflow guidance for using InsideOut infrastructure tools. Call help() for a compact overview, or help(section=...) for a detailed guide. Sections: workflow, tools, examples, inspect. Responses i
Structured diff showing what would be deployed if the user ran tfdeploy now. Returns component-level changes (added/removed/modified), field-level details, and pricing deltas. Defaults (#1392): with n
Create a draft version by reverting to a previous version's config. Copies components, config, and pricing from the target version. If a draft already exists, updates it in-place (single-draft rule).
List all stack versions for a session (newest first). Shows version history including version number, status (draft/confirmed/applied), change summaries, and timestamps. Use this tool to see the desig
FEEDBACK: Submit feedback, bug reports, or feature requests to Luther Systems Use this tool to forward user feedback directly to the Luther Systems team. This includes bug reports, feature requests, q
WORKFLOW: Step 4 of 4 - Deploy infrastructure to the cloud Deploy infrastructure by starting a Terraform job for an InsideOut session. This tool initiates the actual deployment process after Terraform
DESTROY: Tear down previously deployed infrastructure Destroys infrastructure by calling the Oracle destroy endpoint for a session that has a prior successful deployment. IMPORTANT: This starts a long
DRIFT CHECK: Run a read-only drift detection check Checks whether deployed infrastructure has drifted from the expected Terraform state. This is a read-only operation — it does NOT modify any infrastr
WORKFLOW: Step 3 of 4 - Generate Terraform files from completed design Generate Terraform files from an InsideOut session that has completed infrastructure design. ⚠️ PREREQUISITE: Only call this AFTE
MONITORING: Fetch Terraform deployment logs with pagination Fetches logs from a running or completed Terraform deployment job. For **completed jobs**: uses REST endpoint for instant retrieval (support
INSPECTION: Retrieve Terraform outputs from a completed deployment Returns structured output values (VPC IDs, endpoints, cluster names, etc.) after a successful deploy. Sensitive outputs are redacted
PREVIEW: Run terraform plan to preview infrastructure changes Runs a terraform plan for an InsideOut session without applying any changes. This lets the user review what will be created/changed/destro
INSPECTION: List all Terraform deployment runs for a session Returns job IDs, statuses, types (apply/destroy), and timestamps for every run. Use this to see deployment history, find job IDs for log in
MONITORING: Quick status check for Terraform deployments Check the current status of a Terraform deployment job. Use this tool to quickly check if a deployment is running, completed, or failed. Return