Ds Toolkit MCP Server
mcp-ds-toolkit-server
PyPI
v0.3.0
Published by yasserelhaddar — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
Pure standalone MCP Data Science Server - Complete DS workflows with local SQLite tracking, no external services required
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 11 = 89. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −11 | Unsafe deserialization ×3 MTC-SRC-007 | permissions |
2. Client adoption risk — 89 − 4 = 85. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −3 | capability blast radius (moderate) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 5
medium Dynamic module load from a non-literal (src/mcp_ds_toolkit_server/server.py)MTC-SRC-005
In the server's implementation (`src/mcp_ds_toolkit_server/server.py:202`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: __import__(module_name) except ImportError as e: error_msg = f"Criti
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/mcp_ds_toolkit_server/server.py
medium Unsafe deserialization (src/mcp_ds_toolkit_server/training/trainer.py)MTC-SRC-007
In the server's implementation (`src/mcp_ds_toolkit_server/training/trainer.py:721`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: model = pickle.load(f) self.logger.info(f"Model loaded from {model_path}") retu
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/mcp_ds_toolkit_server/training/trainer.py
medium Unsafe deserialization (src/mcp_ds_toolkit_server/utils/model_resolver.py)MTC-SRC-007
In the server's implementation (`src/mcp_ds_toolkit_server/utils/model_resolver.py:132`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: model = pickle.load(f) if isinstance(model, BaseEstimator):
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/mcp_ds_toolkit_server/utils/model_resolver.py
medium Unsafe deserialization (src/mcp_ds_toolkit_server/utils/persistence.py)MTC-SRC-007
In the server's implementation (`src/mcp_ds_toolkit_server/utils/persistence.py:133`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: 8')) return pickle.loads(model_bytes) else: raise ValueError(f"Unsupported encoding for
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/mcp_ds_toolkit_server/utils/persistence.py
low Unconstrained path parameter "output_path" on "export_dataset"MTC-CAP-008
Tool "export_dataset" takes a path parameter "output_path" with no constraint. Without a canonicalize-and-contain check (not visible statically), this permits ../ traversal outside the intended root.
Fix: Resolve and verify the path stays within an allowed root; reject traversal sequences.
Location: tool export_dataset · inputSchema.properties.output_path
Tools 30
Each tool and what it can reach — statically extracted from the published source.
export_datasetreads sensitive databatch_process_datasetsno sensitive capabilityclean_datasetno sensitive capabilityclear_all_datano sensitive capabilitycompare_datasetsno sensitive capabilitycompare_modelsno sensitive capabilitycompare_runsno sensitive capabilitycreate_experimentno sensitive capabilityend_runno sensitive capabilityevaluate_modelno sensitive capability
Show 20 more tools ↓
get_dataset_infono sensitive capabilityget_experimentno sensitive capabilityget_model_infono sensitive capabilitylist_algorithmsno sensitive capabilitylist_datasetsno sensitive capabilitylist_experimentsno sensitive capabilitylist_runsno sensitive capabilityload_datasetno sensitive capabilitylog_artifactno sensitive capabilitylog_metricsno sensitive capabilitylog_paramsno sensitive capabilitypreprocess_datasetno sensitive capabilityprofile_datasetno sensitive capabilityremove_datasetno sensitive capabilitysample_datasetno sensitive capabilitysplit_datasetno sensitive capabilitystart_runno sensitive capabilitytrain_modelno sensitive capabilitytune_hyperparametersno sensitive capabilityvalidate_datasetno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (30 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 1
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v0.3.0 latest |
B 85/100 | 5 | 1.13.0 | 2026-08-25 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan mcp-ds-toolkit-server --online --registry pypiMore in AI & Agents
Authenticated MCP transport with HTTP Signatures for AAuth agents
Local-first MCP server for parallel AI coding agents to claim file ownership before edits, preventing stomping on each other in the same worktree.
Agent-agnostic intercommunication system — sessions, messaging, channels, shared state, and real-time events
MCP server for AI agent task communication and delegation with diagnostic lifecycle visibility
Programmatic add/link/unlink for MCP servers across 23 AI coding agents (Claude Code, Claude Desktop, Cursor, VS Code, Codex, Gemini CLI, Zed, Cline, OpenCode, Goose, Kiro, Windsurf, and more). Functional API with dry-run support.
MCP server layer exposing agent-mesh orchestrator as an MCP agent