Cloudeval AI MCP server

Read from the package source without running it. The installed server may list more.

• agent_profiles_get

  Fetch one Cloudeval Agent Profile by id.

• agent_profiles_list

  List backend-owned Cloudeval Agent Profiles such as Architecture, Cost, Change Reviewer, Evidence Auditor, and Security Reviewer.

• agent_profiles_runwrite action

  Run a Cloudeval Agent Profile against a project through the normal chat stream contract.

• ask

  Ask Cloudeval a one-shot question, optionally scoped to a project and model.

• auth_status

  Return local Cloudeval authentication status.

• billing_invoices

  Fetch subscription invoices, paid top-up history and billing-cycle status for the authenticated account. Use this for payment receipts; use billing_ledger for individual usage charges or billing_usage for aggregates. The service may create missing provider invoice records for already-paid top-ups and persist receipt links. Requires billing read access and explicit billing or all toolset selection.

• billing_ledger

  Inspect individual Cloudeval usage attempts and credit charges for the authenticated account. Use billing_usage for aggregates or billing_invoices for payment receipts. Defaults to the last 30 calendar days; explicit timestamps override the corresponding range bounds. Returns data.items, data.has_more and data.next_cursor; reuse the cursor with the same filters to read the next page. Requires bill

• billing_notifications

  Return Cloudeval billing notifications for the authenticated account.

• billing_plans

  Return Cloudeval billing plan configuration.

• billing_summary

  Return Cloudeval billing entitlement, credit status, and subscription status.

• billing_topup_checkoutwrite action

  Create a checkout session for a selected top-up pack. This is explicit and externally visible.

• billing_topups

  Return available Cloudeval credit top-up packs.

• billing_usage

  Return Cloudeval billing usage summary with date, granularity, and model filters.

• capabilities_get

  Return Cloudeval CLI and MCP capability metadata for agent planning.

• cloudeval_draft_fix

  Return a non-mutating draft-fix proposal for a finding.

• cloudeval_explain_blast_radius

  Fetch Cloudeval graph-neighborhood evidence for a project resource.

• cloudeval_generate_ci_gate

  Generate Cloudeval CI gate files. This MCP tool returns file contents and does not write files.

• cloudeval_get_finding_evidence

  Read evidence for a finding from a Cloudeval IDE run cache.

• cloudeval_get_resource_context

  Return indexed resources for an IaC file or workspace.

• cloudeval_iac_detect

  Detect ARM, Bicep, Terraform, and OpenTofu files in a workspace.

• cloudeval_iac_index

  Index resources and ranges in an IaC file or workspace.

• cloudeval_review_localwrite action

  Run the IDE local review path. This tool indexes resources and returns local support state.

• config_get

  Return one setting from the selected Cloudeval CLI profile.

• config_profiles

  Return Cloudeval CLI config profile names.

• config_show

  Return the selected Cloudeval CLI profile configuration.

• connections_get

  Fetch one Cloudeval connection by id from the authenticated account's connection list.

• connections_list

  List Cloudeval cloud/template connections visible to the authenticated account.

• credentials_createwrite action

  Create a scoped Cloudeval access-key credential. Secret values are redacted unless showSecret is true.

• credentials_inspect

  Inspect a Cloudeval access-key credential by id.

• credentials_list

  List Cloudeval access-key credentials, optionally scoped by project.

• credentials_revoke

  Revoke a Cloudeval access-key credential.

• credentials_templates

  List Cloudeval access-key credential templates.

• doctor

  Return local Cloudeval CLI diagnostic checks.

• identity_get

  Return Cloudeval identity and capability metadata for the active credential.

• models_default_get

  Return the configured default model for the selected CLI profile.

• models_default_set

  Set the configured default model for the selected CLI profile.

• models_list

  List backend-supported Cloudeval models for the active account or access key.

• open_url

  Build a Cloudeval frontend deep link. Optionally open it in the system browser.

• projects_export_diagram

  Export a GraphEditor-rendered architecture or dependency diagram to a local PNG, JPEG, or SVG file for CLI and MCP agents.

• projects_get

  Fetch one Cloudeval project by id from the authenticated account's project list.

• projects_graph_diff

  Compare two retained project graph snapshots.

• projects_graph_get

  Fetch project graph nodes and relationships for automation.

• projects_graph_insights

  Fetch graph intelligence for overview, impact, critical paths, security, cost, or changes.

• projects_graph_sync_runs

  List recent graph-producing sync runs for a project.

• projects_graph_timeline

  List retained graph snapshots for a project.

• projects_list

  List Cloudeval projects visible to the authenticated account.

• projects_overview

  Fetch a Cloudeval project cockpit overview with graph, report, connection, credit, and deep-link metadata for IDE and agent workflows.

• recipes_get

  Fetch one Cloudeval recipe by id.

• recipes_list

  List Cloudeval reusable recipes and their safety metadata.

• recipes_runwrite action

  Run a Cloudeval recipe. Ask/agent recipes consume model credits; explicit side-effect recipes return commands instead of mutating automatically.

• reports_cost

  Fetch the latest normalized Cloudeval cost report for a project.

• reports_download

  Fetch raw, parsed, or formatted report payloads. Optionally write them to a local output path.

• reports_list

  List cost and Well-Architected reports for a project.

• reports_rules

  Return WAF rule findings from the latest Cloudeval WAF report.

• reports_runwrite action

  Submit report generation jobs for a project. This can consume backend compute and credits.

• reports_show

  Fetch one Cloudeval report by id for a project.

• reports_waf

  Fetch the latest normalized Well-Architected report for a project.

• rules_categories

  List cloud validation check categories.

• rules_get

  Show one cloud validation check by id.

• rules_search

  Search cloud validation checks.

• sessions_export

  Return local Cloudeval CLI session history for the selected profile.

• sessions_get

  Return one local Cloudeval CLI session by thread id.

• sessions_list

  List local Cloudeval CLI session history summaries.

• sessions_search

  Search local Cloudeval CLI session titles and messages.

• status

  Return local Cloudeval CLI status and active configuration metadata.

• template_parse

  Parse a local cloud template JSON file. Parameters files are accepted but optional.

• template_testwrite action

  Run local cloud template test checks. Parameters files are accepted but optional.

• template_validate

  Validate a local cloud template JSON file. Parameters files are accepted but optional.

• Code scan17 source files scanned20/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitystatic API keys via environment variables6/15
• Maintenancelast push 21 days ago15/15
• Maintainer identityregistry namespace matches repository owner; GitHub account older than a year9/10

Safety score

77/100

higher than 45% of 1,119 Developer tools servers

Tools

68

the typical Developer tools server has 7

Tools that change things

10%

typical: 5%. Fewer write tools means less an agent can break

GitHub stars

1

more than 50% of 855 Developer tools servers

Downloads a week

89

more than 56% of 436 Developer tools servers with a package

Cloudeval AI: common questions

Is Cloudeval AI MCP server safe?

Mostly: it is graded B (77/100). Read the Cloudeval AI safety report

How do I install Cloudeval AI?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Cloudeval AI need an API key?

Yes. The registry entry asks for CLOUDEVAL_ACCESS_KEY.

Is Cloudeval AI maintained?

The last commit was 54 days ago (2026-09-10). The latest release is v0.38.5.

What can I use instead of Cloudeval AI?

Servers from other publishers that do the same job: TerraVision MCP server, Cloud FinOps Skill & MCP server and ShipStatic MCP server. Compare all Cloudeval AI alternatives.

How does Cloudeval AI compare with other Developer tools MCP servers?

safety score 77/100 (higher than 45% of 1,119 Developer tools servers); tools 68 (the typical Developer tools server has 7); tools that change things 10% (typical: 5%. Fewer write tools means less an agent can break). Full comparison.