@oneie/claude@0.10.0
ONE Claude Code plugin — /do lifecycle, W0-W4 BUILD engine, 7-canon reconcile, ONE substrate via MCP, signals, and rules
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe reviewed credential flow uses the ONE API key for the configured ONE workspace, and the channel stream is on the package’s related channels.one.ie service. The watcher’s generated replies are constrained to a child process with no allowed tools, so the scanner’s credential and execution flags do not establish malicious behavior.
Decision evidence
public snapshot- The plugin runs its announcement hook on user prompts, and that hook requires a configured ONE workspace slug and API key before proceeding.
- The hook announces the session to the configured workspace using the ONE API key.
- The watcher reads ONE credentials and uses them with the configured workspace; its stream request targets channels.one.ie.
- The watcher generates replies with `claude -p` using only HOME and PATH in its child environment and an empty allowed-tools list.
- Replies are posted to the configured workspace using the same ONE API key.
- The credential flow is for the package’s stated ONE workspace integration; the workspace slug and key are supplied through ONE-specific configuration.
- The watcher is started by the announcement hook and runs only when the session marker exists; it has a 12-hour limit and exits when the marker is removed.
- No unrelated credential collection or unrelated collector was identified in the reviewed runtime flow.
Source & flagged code
7 flaggedPackage source references child process execution.
scripts/blocks-manifest.mjsView on unpkg · L22Package source invokes a package manager install command at runtime.
scripts/blocks-manifest.mjsView on unpkg · L192Findings
Affected versions and remediation
This report applies to @oneie/claude@0.10.0.
See all versions of @oneie/claude for other recorded verdicts.
- Review the evidence and your use of @oneie/claude@0.10.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.