io.github.CSOAI-ORG/fda-samd-mcp

Helps map FDA AI/ML SaMD pathways for 510(k), PMA, and De Novo.

Source

Official Registry

Updated

2026-06-13

// securityCaution

Runs locally
Official / curated
Open source
Community-trusted

Summary

This MCP tool comes from the official registry, is open source, and has been updated within the last year, with no clear high-risk red flags in the provided materials. Documentation is very limited and it is known to execute local code, but no credential requirements or remote endpoints are declared, so the overall posture is low to moderate concern.

CredentialsLow risk

The materials explicitly state that no keys or environment variables are required. No API keys, account tokens, or other sensitive credentials are requested, so credential exposure and misuse risk appears low.

Network egressLow risk

The materials explicitly state that there are no remote endpoints. No external service connections or third-party data transfers are declared, so no clear outbound data path is evident from the available information.

Code executionCaution

The objective checks indicate that this tool executes code; this is a normal MCP capability and warrants caution rather than automatic high risk. The materials do not specify which local processes or system capabilities are invoked, so it should be run in a constrained environment and the source should be reviewed.

Data accessCaution

The provided materials do not specify what local files, directories, or other data resources it can read or write, so the data access boundary is unclear. There is no explicit sign of overbroad permissions, but as an executable MCP tool it should still be isolated under least-privilege principles.

Supply chainCaution

Positive factors include the official registry source, open-source availability, and updates within the last year, all of which materially reduce supply-chain risk. Caution remains because there is no README, no declared license, and community adoption is very low (0 stars), so while the code is auditable, external validation and usage history are limited.

Recommendations

Review the GitHub source and dependency list first to verify actual execution logic, file access scope, and whether any undeclared network requests exist.
Run it in a container, sandbox, or least-privileged account, and avoid granting unnecessary local file access.
Pin a specific version or commit before deployment to reduce supply-chain uncertainty from future upstream changes.
If used for medical compliance-related work, treat it as an assistive tool and keep critical conclusions subject to human review.

Audited by: gpt-5.4 · 2026-06-17

// install

Copy the install command and let the AI configure it · recommended for beginners

Please install the "io.github.CSOAI-ORG/fda-samd-mcp" MCP server from askskill:
Run: claude mcp add 'io-github-csoai-org-fda-samd-mcp' -- npx -y fda-samd-mcp