fail_step

THE RISK

Medium Risk

Why fail_step needs a policy

This tool modifies the state of a plan step by marking it as failed. This is a reversible state change (the step could potentially be retried or reset), making it a Write operation. It doesn't execute code, delete data irreversibly, or involve financial transactions. Misuse could disrupt workflow execution but is recoverable.

From the tool's definition Mark a step as failed with error details

Documented attack patterns abuse exactly the kind of access fail_step gives an agent:

POLICY

How to control fail_step

PolicyLayer is an MCP gateway — it sits between your AI agents and Context Engine MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for fail_step:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "fail_step": {
      "limits": [
        {
          "counter": "fail_step_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

fail_step stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Context Engine MCP Server — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Context Engine MCP Server tools

Destructive clear_index Remove saved index state and clear caches without rebuilding. Destructive delete_plan Delete a saved plan from storage. Destructive reindex_workspace Clear current index state and rebuild it from scratch. Execute execute_plan Execute steps from an implementation plan, generating code changes. This tool orchestrate Execute index_workspace Index the current workspace for semantic search. This tool scans all source files in the Execute reactive_review_pr Start a reactive PR code review session. This tool initiates an AI-powered code review wi Execute run_static_analysis Run local static analyzers (tsc and optional semgrep) and return structured findings. Execute check_invariants Run YAML invariants deterministically against a unified diff (no LLM).

All 50 Context Engine MCP Server tools →

Write tools on other servers

M-Team MCP Server download_torrent YouTube MCP Server generate_video_title Android Forensics ADB MCP Server combine_reports 0xarchive web3_signup

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Argument validation → Checking tool call arguments against schema and policy before they run.
Scoped token → Per-person credentials that grant a defined subset of servers and tools.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about fail_step

What does the fail_step tool do? +

Mark a step as failed with error details. It is categorised as a Write tool in the Context Engine MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on fail_step? +

Register the Context Engine MCP Server MCP server in PolicyLayer and add a rule for fail_step: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Context Engine MCP Server. Nothing to install.

What risk level is fail_step? +

fail_step is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit fail_step? +

Yes. Add a rate_limit block to the fail_step rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block fail_step completely? +

Set action: deny in the PolicyLayer policy for fail_step. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides fail_step? +

fail_step is provided by the Context Engine MCP Server MCP server (kirachon/context-engine). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Context Engine MCP Server tool call.

Start from Context Engine MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →