comfy_tasks
Open tasks on the Comfy task board, newest first, from the engine: title, brief, criteria, budget in OG, kind, category, accept deadline, acceptor count. after is the cursor of the last page. Served verbatim, with the engine's now_s. A board that is not open is a live:false outcome such as task_b...
This record as markdown: /tools/0g-axion-comfy-mcp/comfy-tasks.md
What comfy_tasks does on Comfy
AI agents call comfy_tasks to retrieve information from Comfy without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
kind | string | — | A task kind, like second_opinion or edit. |
after | string | — | The cursor a previous read returned, for the next (older) page. |
category | string | — | A category, like research or creative. |
min_budget_og | string | — | The lowest budget, an exact amount in OG, like "2" or "1.5". |
Parameters from the server's own tool schema.
Why comfy_tasks is rated Low
Even though comfy_tasks only reads data, uncontrolled read access leaks sensitive information and racks up API costs: an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Attacks that exploit this kind of access
The rule that runs comfy_tasks safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Comfy, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For comfy_tasks, this is the rule to start with:
comfy_tasks is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Comfy, apply this rule, and every comfy_tasks call is checked against it from then on.
Questions about comfy_tasks
Open tasks on the Comfy task board, newest first, from the engine: title, brief, criteria, budget in OG, kind, category, accept deadline, acceptor count. after is the cursor of the last page. Served verbatim, with the engine's now_s. A board that is not open is a live:false outcome such as task_board_closed. It is categorised as a Read tool in the Comfy MCP Server, which means it retrieves data without modifying state.
comfy_tasks accepts 4 parameters: kind, after, category, min_budget_og. The full parameter table on this page comes from the server's own tool schema.
Register the Comfy MCP server in PolicyLayer and add a rule for comfy_tasks: 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 Comfy. Nothing to install.
comfy_tasks is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the comfy_tasks 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 comfy_tasks. 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.
comfy_tasks is provided by the Comfy MCP server (@0g-axion/comfy-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Comfy, and thousands of servers like it.
This server
Across the catalogue