Session lifecycle and narrative context for AI game mastering. 🎮 SESSION WORKFLOW: 1. initialize - Start/resume session (loads or creates world + party) 2. get_context - Get comprehensive context for AI decision-making 📋 CONTEXT INCLUDES: - Party members with HP, level, class - Active quests an...
AI agents use session_manage to create or update resources in Rpg — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Rpg environment.
The tool has two actions: 'get_context' is clearly a Read operation, but 'initialize' creates or loads persistent game state (world + party) in a SQLite-backed store, which constitutes a Write (creating new data or modifying session state). Since the most severe applicable category wins, the tool is classified as Write.
From the tool's definition 'initialize - Start/resume session (loads or creates world + party)' and 'get_context - Get comprehensive context for AI decision-making'
Documented attack patterns abuse exactly the kind of access session_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 session_manage:
{
"version": "1",
"default": "deny",
"tools": {
"session_manage": {
"limits": [
{
"counter": "session_manage_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}session_manage 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.
Session lifecycle and narrative context for AI game mastering. 🎮 SESSION WORKFLOW: 1. initialize - Start/resume session (loads or creates world + party) 2. get_context - Get comprehensive context for AI decision-making 📋 CONTEXT INCLUDES: - Party members with HP, level, class - Active quests and current objectives - World state (time, location, weather) - Recent narrative events - Active combat status 💡 AI USAGE: Call get_context at conversation start to understand game state. Inject context into system prompt for informed storytelling. Actions: initialize, get_context Aliases: init/start→initialize, context/narrative→get_context. It is categorised as a Write tool in the Rpg MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Rpg MCP server in PolicyLayer and add a rule for session_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.
session_manage 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 session_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 session_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.
session_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.