Skip to main content
Glama
ComplianceCow

ComplianceCow MCP Server

execute_task

Execute compliance tasks using real data, automatically resolving dependencies by executing prerequisite tasks first and using their outputs as inputs.

Instructions

Execute a specific task with real data after collecting all required inputs.

**This tool executes tasks with REAL data, not sample data.**
If any input depends on a previous task's output and that output is not available,
the dependent task(s) MUST be executed first to obtain the real output.

===============================================================================
EXECUTION CONTEXT
===============================================================================
- This tool MUST be called after collecting the inputs for a task.
- Execution is sequential: execute Task 1 → then Task 2 → etc.
- No task may proceed until its dependent tasks have been executed.
- On execution failure, provide detailed error feedback.

===============================================================================
DEPENDENCY & REAL DATA HANDLING
===============================================================================
If a task requires input from a previous task (dataset, file, or structured output):

1. **Use real task output when available**
    - If the dependent task was already executed and produced outputs:
        → Use those outputs as the input.
        → Do NOT generate synthetic/sample data.
        → Do NOT re-run the previous task unnecessarily.

2. **If required previous task output does NOT exist**
    - The assistant MUST:
        - Explain *why* execution of the previous task is required.
        - Automatically execute the previous task (and any required tasks in the chain).
        - **After execution, display all execution results and outputs.**
        - NO user confirmation should be requested—only explanation.
        - Use the REAL output from the executed task as input.

3. **If executing a required previous task fails**
    - The assistant MUST:
        - Explain clearly why the task failed.
        - Ask the user to provide the required input data manually.
    - User-provided data becomes the fallback input.

4. **Only execute what is needed**
    - Execute ONLY the minimal set of tasks whose outputs are required.
    - **Every executed task must have its results shown to the user immediately.**

===============================================================================
APPLICATION CONFIGURATION
===============================================================================
Application credentials are REQUIRED if the task's appType is NOT 'nocredapp'.

If the task requires application credentials (appType != 'nocredapp'):
- Application config must be provided with:
    - appName: Application class name
    - appURL: Application URL (optional, can be empty string)
    - credentialType: Type of credentials
    - credentialValues: Actual credential key-value pairs
- OR applicationId if using existing saved application

If the task's appType is 'nocredapp':
- Application configuration can be omitted (pass None or empty)
- The system will automatically use the hardcoded nocredapp application structure:
  {
      "applicationType": "NoCredApp",
      "appURL": "",
      "credentialType": "NoCred",
      "credentialValues": {"Dummy": ""},
      "appTags": {"appType": ["nocredapp"], "environment": ["logical"], "execlevel": ["app"]}
  }

===============================================================================
TASK EXECUTION FLOW
===============================================================================
1. Receive task name and collected inputs
2. Check if any input depends on previous task output
3. For dependency inputs:
    a. Check if previous task output exists
    b. If not, execute previous task first
    c. Use real output as input value
4. Prepare execution payload with real data
5. Call task execution API
6. Parse and return execution results with output file URLs

===============================================================================
REQUEST BODY FORMAT
===============================================================================
    {
        "taskname": "TaskName",
        "application": {
            "appName": "ApplicationClassName",
            "appURL": "https://app.url.com",
            "credentialType": "CredentialTypeName",
            "credentialValues": {
                "key1": "value1",
                "key2": "value2"
            },
            "appTags": [Complete object from of 'appTags' from the task in the rule]
        },
        "taskInputs": {
            "inputs": {
                "InputName1": "value_or_file_url",
                "InputName2": "value_or_file_url"
            }
        }
    }
===============================================================================
Args:
    task_name: Name of the task to execute
    task_inputs: Dictionary containing key-value pairs of task inputs
                Format: {"input_name": "value" or file_url}
    application: Optional application configuration for tasks requiring credentials
                Format: {
                    "appName": "ApplicationClassName",
                    "appURL": "https://...",
                    "credentialType": "...",
                    "credentialValues": {...},
                    "appTags": [Complete object from of 'appTags' from the task in the rule]
                }
                OR {"applicationId": "existing-app-id", "appTags": [Complete object from of 'appTags' from the task in the rule]}

Returns:
    Dict containing:
    {
        "success": bool,
        "execution_status": "COMPLETED" | "FAILED",
        "task_name": str,
        "task_inputs": dict,
        "outputs": dict,  # Output file URLs and values
        "errors": list,
        "message": str,
        "next_action": str
    }

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
task_nameYes
task_inputsYes
applicationNo

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Given no annotations, the description carries full behavioral burden. It clearly states the tool executes with real data, handles dependencies, and shows results. However, it could explicitly mention that execution may trigger real-world actions (e.g., data modification) beyond what the name implies, but the emphasis on 'real data' and the execution flow strongly implies mutation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is lengthy but well-structured with clear sections (execution context, dependency handling, configuration, flow) and bullet points. It is front-loaded with the core purpose. While every sentence adds value, it could be slightly more concise, but the complexity warrants the length.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description covers all aspects: input parameters, dependency logic, error handling, application configuration, execution flow, and return value format. Given the tool's complexity and the presence of an output schema, it is complete and leaves no significant gaps for an AI agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 0% schema description coverage, the description fully compensates by detailing parameter formats: task_name as a string, task_inputs as key-value pairs, and application with appName, credentialType, etc. It provides a complete request body example and explains optionality based on appType, adding critical meaning beyond the minimal schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: executing a specific task with real data after input collection. It emphasizes real data vs. sample, sequential execution, and dependency handling, distinguishing it from sibling tools like execute_rule or fetch_task_details.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit guidance on when to use the tool ('MUST be called after collecting inputs'), how to handle dependencies (execute previous tasks automatically, use real outputs), and when not to use it (do not generate synthetic data). It also explains fallback on failure and the minimal execution principle.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ComplianceCow/cow-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server