Send a photo to a chat.
AI agents use send_photo to create or update resources in Telegram Mcp Kit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Telegram Mcp Kit environment.
send_photo writes/creates new data (a photo message) to a Telegram chat. While messages can technically be deleted later, the primary action is content creation, placing it in the Write category rather than Execute (which would apply if the tool triggered arbitrary external operations).
From the tool's definition Tool description states 'Send a photo to a chat' — this creates new content (a message with photo attachment) in a chat, modifying the chat's message history irreversibly from the perspective of undo operations, but the action is not destructive in the sense…
Attacks that exploit this kind of access
Send a photo to a chat. It is categorised as a Write tool in the Telegram Mcp Kit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Telegram Mcp Kit MCP server in PolicyLayer and add a rule for send_photo: 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 Telegram Mcp Kit. Nothing to install.
send_photo 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 send_photo 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 send_photo. 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.
send_photo is provided by the Telegram Mcp Kit MCP server (quoctang/telegram-mcp-kit). 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 →