Bigpictureio
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill is a broad Membrane-powered Bigpicture.io integration, but its purpose is inconsistent and it gives the agent wide authenticated action-running ability through unreviewed/generated tooling.
Install only if you trust Membrane and can verify this is the correct Bigpicture.io integration. Before use, confirm which Bigpicture.io service/account is being connected, avoid auto-built connectors unless reviewed, and require explicit approval before the agent performs write, admin, billing, permission, token, webhook, or security-related actions.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user could connect the wrong account or approve actions under a mistaken understanding of the integration's purpose.
The stated service purpose conflicts with the listed common actions, which makes it unclear what the skill is actually meant to do and what data/account the user is approving.
BigPicture.io is a project management tool... Popular actions... Find Company by IP Address... Search Company by Name
Clarify whether this is for BigPicture project management or BigPicture company/IP enrichment, and remove unrelated resource/action descriptions.
If the agent selects a mutating or administrative action, it could change business records or account settings without a clearly documented confirmation boundary.
The skill delegates broad action discovery and execution to the agent through an authenticated connection, but does not define safe action classes, approval requirements, or limits for mutating operations.
Use action names and parameters as needed... membrane action list --connectionId=CONNECTION_ID --intent "QUERY"... membrane action run <actionId> --connectionId=CONNECTION_ID --input '{"key": "value"}'Require explicit user confirmation before create/update/delete, billing, permission, token, webhook, or security-related actions, and prefer read-only actions unless the user asks for a change.
The user may be granting credentials and data access to code or connector behavior that was not statically reviewable in this skill package.
The setup depends on a globally installed latest-version CLI and may rely on an automatically generated connector that is not included in the reviewed artifacts.
npm install -g @membranehq/cli@latest... If no app is found, one is created and a connector is built automatically.
Pin the CLI version, provide a reviewed connector/action manifest, and require user approval before using an auto-built connector.
The skill can act through the user's authenticated Membrane/Bigpicture.io connection until access is revoked or expires.
Delegated account access and credential refresh are expected for a SaaS integration, but they are sensitive capabilities that users should consciously approve.
Membrane handles authentication and credentials refresh automatically... membrane login --tenant --clientName=<agentType>
Use the least-privileged account available, review requested authorization screens, and know where to revoke the Membrane/Bigpicture.io connection.
Business data sent to or returned from Bigpicture.io may also be processed by Membrane as part of the integration flow.
Membrane acts as an intermediary/gateway for authentication and action execution, so user inputs and action outputs may pass through that provider.
This skill uses the Membrane CLI to interact with Bigpicture.io. Membrane handles authentication and credentials refresh automatically
Confirm Membrane is an approved provider for the data involved and avoid sending highly sensitive records unless the provider relationship is acceptable.