Mark the plan as failed. Call this if an unrecoverable error occurs during execution.
AI agents use plan_failed to create or update resources in Overture — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Overture environment.
This tool creates or modifies data reversibly—it changes the plan's status flag to indicate failure. This is not destructive (the plan data remains; the status is just updated), not financial, and not execute (it doesn't run code or trigger external operations). The modification is reversible (the status could theoretically be updated again).
From the tool's definition The tool description states 'Mark the plan as failed,' which is a state modification operation. The tool modifies the status/state of a plan object from its current state to a 'failed' state.
Documented attack patterns abuse exactly the kind of access plan_failed gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Overture, and nothing reaches the server without passing your rules. This is the rule we recommend for plan_failed:
{
"version": "1",
"default": "deny",
"tools": {
"plan_failed": {
"limits": [
{
"counter": "plan_failed_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}plan_failed 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.
Free to start. No card required.
Mark the plan as failed. Call this if an unrecoverable error occurs during execution. It is categorised as a Write tool in the Overture MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Overture MCP server in PolicyLayer and add a rule for plan_failed: 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 Overture. Nothing to install.
plan_failed is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the plan_failed 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.
Set action: deny in the PolicyLayer policy for plan_failed. 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.
plan_failed is provided by the Overture MCP server (sixhq/overture). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 14 Overture tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
14 Overture tools catalogued and risk-classified — across an index of 42,500+ MCP servers.