Manage improvised actions, custom effects, and arcane synthesis. Actions: stunt, apply_effect, get_effects, remove_effect, process_triggers, advance_durations, synthesize, get_spellbook Aliases: rule_of_cool->stunt, boon/curse->apply_effect, dispel->remove_effect, arcane_synthesis->synthesize STU...
AI agents invoke improvisation_manage to trigger actions in Rpg. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes game-engine logic with real side effects on persistent SQLite-backed state: applying/removing effects, processing trigger chains, advancing durations, and synthesizing new spells. It spans Write and Execute; the trigger processing and arcane synthesis represent dynamic execution of game mechanics with unpredictable cascading outcomes, placing it firmly in Execute.
From the tool's definition 'apply_effect', 'remove_effect', 'process_triggers', 'advance_durations', 'synthesize' — triggers game-engine operations, modifies active state, processes cascading trigger chains, and synthesizes new arcane effects; 'stunt' can cause self-damage on critical…
Documented attack patterns abuse exactly the kind of access improvisation_manage gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Rpg, and nothing reaches the server without passing your rules. This is the rule we recommend for improvisation_manage:
{
"version": "1",
"default": "deny",
"tools": {
"improvisation_manage": {
"limits": [
{
"counter": "improvisation_manage_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}improvisation_manage stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Manage improvised actions, custom effects, and arcane synthesis. Actions: stunt, apply_effect, get_effects, remove_effect, process_triggers, advance_durations, synthesize, get_spellbook Aliases: rule_of_cool->stunt, boon/curse->apply_effect, dispel->remove_effect, arcane_synthesis->synthesize STUNT (Rule of Cool): - DC 5-30 based on difficulty - Supports advantage/disadvantage - Critical success doubles damage - Critical failure can cause self-damage CUSTOM EFFECTS: - Categories: boon, curse, neutral, transformative - Power levels 1-5 - Duration types: rounds, minutes, hours, days, permanent ARCANE SYNTHESIS: - DC = 10 + (spell level x 2) + modifiers - Outcomes: mastery (learned!), success, fizzle, backfire, catastrophic - Mastery permanently adds spell to spellbook. It is categorised as a Execute tool in the Rpg MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Rpg MCP server in PolicyLayer and add a rule for improvisation_manage: 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 Rpg. Nothing to install.
improvisation_manage is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the improvisation_manage 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 improvisation_manage. 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.
improvisation_manage is provided by the Rpg MCP server (mnehmos/mnehmos.rpg.mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Rpg, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
47 Rpg tools catalogued and risk-classified — across an index of 43,000+ MCP servers.