paperdebugger

The materials and objective checks indicate no required keys or environment variables. No API tokens, account credentials, or other high-sensitivity secrets are requested, so credential exposure and abuse risk appears low.

The current materials list no remote host endpoints, and the README does not disclose any external service connections or data upload paths. Based on available facts, no explicit network egress is evident, though documentation transparency is limited.

The objective checks explicitly mark executes-code, indicating the tool can spawn local processes or execute code. This is a common MCP/developer-tool capability and does not by itself imply high risk, but it should be used in a constrained environment with its callable scope reviewed.

As a plugin-based multi-agent system for in-editor academic writing, review, and editing, it likely needs access to editor text, paper drafts, or related local files. The materials do not define precise read/write boundaries; there is no clear evidence of excessive permissions, but the data-access scope is not transparent and should be minimized.

The source is an open GitHub repository under AGPL-3.0 with about 1.5k stars, providing meaningful visibility and auditability; these are positive signals that reduce risk. The main gaps are the missing README and unknown maintenance status, so code, release artifacts, and dependency pinning should still be reviewed before adoption.