Consequence evaluation for AI agents. Check recoverability before destructive actions.
Consequence layer for AI agents Check recoverability before destructive actions Recourse is an MCP server that evaluates Terraform plans, shell commands, and tool calls before execution. It returns structured facts — recoverability tier, evidence assessment, and risk level — so callers can make context-aware decisions. Agents call Recourse before they act; humans see what the agent checked. One…
Inferred from the transports this listing declares (stdio). A client not listed here hasn’t been ruled out — it just isn’t something Forge can confirm.
Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts.
Read out of the source npm actually ships, at scan time. The package was never executed. Tools registered dynamically at runtime, or hidden inside bundled or minified code, can be missed — so this is a floor on the tool surface, not a complete census of it.
recourse_gateway_statusNo description publishedThis tool published no description. Forge does not invent one.
gateway_terraform_planNo description publishedThis tool published no description. Forge does not invent one.
gateway_terraform_applyNo description publishedThis tool published no description. Forge does not invent one.
gateway_terraform_destroyprivilegedNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_getNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_logsNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_describeNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_applyNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_deleteprivilegedNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_scaleNo description publishedThis tool published no description. Forge does not invent one.
gateway_kubectl_execprivilegedNo description publishedThis tool published no description. Forge does not invent one.
gateway_shell_execprivilegedNo description publishedThis tool published no description. Forge does not invent one.
gateway_request_approvalNo description publishedThis tool published no description. Forge does not invent one.
gateway_check_approvalNo description publishedThis tool published no description. Forge does not invent one.
gateway_get_planNo description publishedThis tool published no description. Forge does not invent one.
recourse_evaluate_terraformprivilegedNo description publishedThis tool published no description. Forge does not invent one.
recourse_evaluate_shellprivilegedNo description publishedThis tool published no description. Forge does not invent one.
recourse_evaluate_mcp_callprivilegedNo description publishedThis tool published no description. Forge does not invent one.
recourse_supported_resourcesNo description publishedThis tool published no description. Forge does not invent one.
recourse_evaluate_with_evidenceprivilegedNo description publishedThis tool published no description. Forge does not invent one.
recourse_agent_instructionsNo description publishedThis tool published no description. Forge does not invent one.
0 of 21 tools published a description.
Tool names and descriptions are written by the publisher and shown verbatim as inert text. They are the strings an MCP client passes to a model, so Forge scans them for prompt-injection patterns — any finding appears with the security scan above. “Privileged” is a keyword match on the tool name, not an audit of what the tool does: a benign-sounding name can still do anything.
Consequence layer for AI agents Check recoverability before destructive actions Recourse is an MCP server that evaluates Terraform plans, shell commands, and tool calls before execution. It returns structured facts — recoverability tier, evidence assessment, and risk level — so callers can make context-aware decisions. Agents call Recourse before they act; humans see what the agent checked. One config block. Works with Claude Desktop, Claude Code, Cursor, and any MCP-compatible client. The…
Linked names open Forge’s index of every entry observed exposing that tool. Browse all indexed tools.
This package was last scanned before Forge began storing the resolved tree. The next scan will record it.