Ulysses Protocol: surprise-gated debugging for autonomous agents. Forces pre-committed recovery, rigorous surprise assessment, and falsifiable hypotheses. Operations: - init: Start a debugging session (args: { problem, constraints? }) - plan: Record primary action + pre-committed recovery step (a...
Part of the Thoughtbox server.
Free to start. No card required.
AI agents may call thoughtbox_ulysses to permanently remove or destroy resources in Thoughtbox. Without a policy, an autonomous agent could delete critical data in a loop with no way to undo the damage. PolicyLayer blocks destructive tools by default and requires explicit human approval before enabling them.
Without a policy, an AI agent could call thoughtbox_ulysses in a loop, permanently destroying resources in Thoughtbox. There is no undo for destructive operations. PolicyLayer blocks this tool by default and only allows it when a human explicitly approves the action.
Destructive tools permanently remove data. Block by default. Only enable with explicit approval workflows.
{
"version": "1",
"default": "deny",
"hide": [
"thoughtbox_ulysses"
]
}See the full Thoughtbox policy for all 85 tools.
These attack patterns abuse exactly the kind of access thoughtbox_ulysses gives an agent. Each links to the full case and the policy that stops it:
Other destructive tools across the catalogue. The same approach applies to each: deny by default, or require human approval.
Ulysses Protocol: surprise-gated debugging for autonomous agents. Forces pre-committed recovery, rigorous surprise assessment, and falsifiable hypotheses. Operations: - init: Start a debugging session (args: { problem, constraints? }) - plan: Record primary action + pre-committed recovery step (args: { primary, recovery, irreversible? }) - outcome: Assess step result and update surprise state (args: { assessment, severity?, details? }) - reflect: Form falsifiable hypothesis when S=2 (args: { hypothesis, falsification }) - status: Show current session state (S register, active step, surprise register) - complete: End session with terminal status (args: { terminalState:. It is categorised as a Destructive tool in the Thoughtbox MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Thoughtbox MCP server in PolicyLayer and add a rule for thoughtbox_ulysses: 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 Thoughtbox. Nothing to install.
thoughtbox_ulysses is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the thoughtbox_ulysses 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 thoughtbox_ulysses. 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.
thoughtbox_ulysses is provided by the Thoughtbox MCP server (@kastalien-research/thoughtbox). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 85 Thoughtbox tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.