Loading ToolBench

Reading the current corpus…

genpark-agent-artifact-cache-version-checkpoint-skill MCP server · F · ToolBench

← Browse

genpark-agent-artifact-cache-version-checkpoint-skill

github.com/alphaparkinc/genpark-agent-artifact-cache-version-checkpoint-skill

MCP server for managing agent artifact cache checkpoints with version control capabilities

2026-07-28+STDIO1 tools

20/100

Definition Quality18

Protocol Readiness15

Supportability25

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

This server is non-functional. The MCP server implementation is a stub that prints a JSON status message and exits. The single declared tool 'execute_skill_action' has no input schema, no parameter definitions, and no actual implementation. The description states the tool is 'not implemented in provided source.' The client.py and example_usage.py files are disconnected from the MCP server and do not integrate with it. No tool registration, no schema validation, no error handling, and no protocol compliance.

Tools (1)

execute_skill_actionwritesource verified18/100

Execute a skill action (declared in supported_tools but not implemented in provided source)

Issues (5)

CRITICALtool ↗

No input schema defined for execute_skill_action. Tool has zero parameter definitions, types, or constraints.

execute_skill_action

CRITICALtool-description ↗

Tool description is incomplete and admits non-implementation: 'declared in supported_tools but not implemented in provided source.' LLMs cannot determine when or why to select this tool. Violates pattern:tool-description baseline (10 - 1024 chars, non-empty).

execute_skill_action

CRITICALmcp-server ↗

MCP server implementation is a stub. run_mcp_server() prints a JSON status and exits. No request/response loop, no tool invocation handler, no protocol state machine. Server cannot receive or process tool calls.

execute_skill_action

CRITICALtool-chain ↗

Client code (client.py, example_usage.py) is disconnected from MCP server. AgentArtifactCacheVersionCheckpointClient is a standalone class with no MCP integration. The checkpoint creation logic does not wire into the MCP tool.

Recommendations

Implement a proper MCP server request/response loop using the MCP SDK (e.g., mcp.Server). The current stub prints JSON once and exits.
Define a complete JSON Schema for execute_skill_action input parameters. Include parameter names, types (string, number, boolean, object, array), descriptions, and constraints (enum, minLength, pattern, etc.).
Expand the tool description to 50 - 200 characters explaining WHAT the tool does, WHEN to use it, and WHAT it returns. Example: 'Execute a skill action by name and parameters. Returns execution status, result data, and any errors. Use this to trigger custom agent behaviors.'
Integrate the AgentArtifactCacheVersionCheckpointClient logic into the MCP tool handler. Map the 'execute_skill_action' call to the checkpoint creation workflow, or split into separate tools (create_checkpoint, get_checkpoint_status).
Add error handling that returns structured error responses with actionable guidance. Example: '{"error": "skill_not_found", "message": "Skill 'xyz' not found. Available skills: [list]", "retryable": false}'.

Score history

Overall score trend

First recorded score · v2 rubric

20/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	20	2026-07-28+	v2