upload_resource

// WHEN AI AGENTS USE THIS TOOL

AI agents use upload_resource to create or modify resources in Ai Agent. Write operations carry medium risk because an autonomous agent could trigger bulk unintended modifications. Rate limits prevent a single agent session from making hundreds of changes in rapid succession. Argument validation ensures the agent passes expected values.

// WHY ENFORCE A POLICY ON UPLOAD_RESOURCE

Without a policy, an AI agent could call upload_resource repeatedly, creating or modifying resources faster than any human could review. Intercept's rate limiting ensures write operations happen at a controlled pace, and argument validation catches malformed or unexpected inputs before they reach Ai Agent.

// RECOMMENDED POLICY

Write tools can modify data. A rate limit prevents runaway bulk operations from AI agents.

ai-agent.yaml

tools:
  upload_resource:
    rules:
      - action: allow
        rate_limit:
          max: 30
          window: 60

See the full Ai Agent policy for all 8 tools.

// DETAILS

Tool Name upload_resource

Category Write

MCP Server Ai Agent MCP Server

Risk Level Medium

// MORE AI AGENT TOOLS

Destructive uninstall_resource Execute sync_resources Write manage_subscription Write resolve_prompt_content Write track_usage Read query_usage_stats Read search_resources

// WHAT CAN GO WRONG

Agents calling write-class tools like upload_resource have been implicated in these attack patterns. Read the full case and prevention policy for each:

Browse the full MCP Attack Database →

// OTHER WRITE TOOLS ACROSS MCP SERVERS

Other tools in the Write risk category across the catalogue. The same policy patterns (rate-limit, validate) apply to each.

AbraFlexi contact_create AbraFlexi contact_update AbraFlexi evidence_create AbraFlexi evidence_update AbraFlexi invoice_issued_update AbraFlexi invoice_received_create

// RELATED READING

// FAQ

What does the upload_resource tool do? +

Upload a new AI resource (command, skill, rule, or mcp) to a CSP source repository. The user selects files from their local machine — read each file and pass its content in files[]. ALWAYS confirm the target source repo with the user (e.g. "csp" (default) or "client-sdk-ai-hub"). Resource type rules: • If the user explicitly states the type, use it as-is — no overriding. • If the user does NOT state a type, auto-detect from file structure: - Contains mcp-config.json → type="mcp" - Contains SKILL.md → type="skill" - Single .mdc file → type="rule" - Single .md file → type="command" • If the user says type="mcp" but mcp-config.json is missing, the tool will return an error with a hint about creating mcp-config.json. Resource name rules: • If the user provides a name, use it. • Otherwise derive the name from the filename WITHOUT its extension. Keep the original filename — NEVER rename files (e.g. do not rename any .md file). Examples: "code-review.md" → name="code-review"; "code-review/SKILL.md" → name="code-review". Pass files[] — an array of {path, content} entries. path must be the original filename as-is (relative, no path traversal). No restriction on file extensions — mcp packages may include .py, .js, package.json, etc. The user_token is injected automatically by the server — do NOT read or pass it manually.. It is categorised as a Write tool in the Ai Agent MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on upload_resource? +

Add a rule in your Intercept YAML policy under the tools section for upload_resource. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Ai Agent MCP server.

What risk level is upload_resource? +

upload_resource is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit upload_resource? +

Yes. Add a rate_limit block to the upload_resource rule in your Intercept 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.

How do I block upload_resource completely? +

Set action: deny in the Intercept policy for upload_resource. 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.

What MCP server provides upload_resource? +

upload_resource is provided by the Ai Agent MCP server (@elliotding/ai-agent-mcp). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Let agents act without letting them run wild.