Bugbug
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This appears to be a legitimate BugBug/Membrane integration, but it gives the agent broad authenticated API proxy power, including update and delete methods, without clear approval limits.
Install only if you are comfortable connecting BugBug through Membrane. Prefer the listed BugBug actions, use least-privileged credentials, and require explicit approval before any raw proxy request or any operation that updates, stops, runs, or deletes BugBug resources.
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.
If misused, the agent could make broad authenticated changes to BugBug data, such as modifying or deleting resources, without the skill itself requiring an explicit confirmation step.
This gives the agent a raw authenticated API escape hatch, including mutation and delete methods, instead of limiting it to pre-scoped actions.
you can send requests directly to the BugBug API through Membrane's proxy ... injects the correct authentication headers ... HTTP method (GET, POST, PUT, PATCH, DELETE)
Use named actions where possible, require explicit user approval before POST/PUT/PATCH/DELETE or raw proxy calls, and show the exact endpoint, method, and body before execution.
Connecting the skill grants Membrane-backed access to your BugBug account within whatever permissions that account has.
The skill relies on delegated account credentials and automatic refresh, which is expected for this integration but sensitive.
Membrane handles authentication and credentials refresh automatically
Use the least-privileged BugBug/Membrane account available and revoke the connection when no longer needed.
The installed CLI version may change over time and was not part of the static artifact scan.
The setup depends on installing the latest global npm CLI package, which is purpose-aligned but not version-pinned in the reviewed artifacts.
npm install -g @membranehq/cli@latest
Install from the official Membrane source, consider pinning a reviewed CLI version, and keep the CLI updated intentionally.
A remote response could influence how the agent proceeds during connection setup.
Provider-returned setup data may include instructions directed at the agent, which should be treated as contextual guidance rather than higher-priority instructions.
`clientAction.agentInstructions` (optional) — instructions for the AI agent on how to proceed programmatically
Do not let returned agentInstructions override the user's request, safety rules, or confirmation requirements.
Request paths, request bodies, and API responses may pass through Membrane while interacting with BugBug.
BugBug API traffic is routed through Membrane as a proxy/gateway, which is disclosed and expected but relevant for data-flow awareness.
send requests directly to the BugBug API through Membrane's proxy
Avoid sending unnecessary sensitive data in raw proxy requests and review Membrane's data handling policies before connecting.