Снять права админа (все флаги в False).
AI agents use demote_admin to create or update resources in Claudegram — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claudegram environment.
This tool removes admin privileges from a user in a Telegram chat/group. It modifies user permissions reversibly (admin rights can be re-granted), placing it in the Write category. However, the blast radius is high because demoting an admin in a group can disrupt group management, lock out legitimate admins, and cause significant operational harm if misused by an AI agent.
From the tool's definition Снять права админа (все флаги в False) — 'demote admin (all flags to False)'
Attacks that exploit this kind of access
Снять права админа (все флаги в False). It is categorised as a Write tool in the Claudegram MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claudegram MCP server in PolicyLayer and add a rule for demote_admin: 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 Claudegram. Nothing to install.
demote_admin 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 demote_admin 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 demote_admin. 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.
demote_admin is provided by the Claudegram MCP server (sanjar-x/claudegram). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →