@telora/daemon@0.23.54
Agent orchestration daemon for Telora - spawns and manages Claude Code instances
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe source supports a warning for first-party setup of a workstation-wide Codex instruction file, whose scope extends beyond the connected repository. The inspected runtime-operation hints concern package provisioning, install-input analysis, and hot reload, and do not establish malicious behavior.
Decision evidence
public snapshot- The manifest defines a prepublish build hook, not an install hook.
- The Codex connector writes a Telora instruction seed to the workstation-wide Codex AGENTS.md; daemon startup refreshes it when present.
- The seed merge appends or replaces Telora’s delimited block, preserving surrounding instructions.
- The reviewed worktree installer, install-input scanner, and hot-reload code perform package setup, analysis, and configuration reload operations; the flagged runtime locations do not establish an attack.
- The connector files describe harness-specific setup, including Claude settings, Codex MCP configuration, and IDE MCP registration.
- The global Codex seed is installed through the explicit harness connection flow; the startup refresh is limited to a seed file that already exists.
- No credential disclosure, destructive behavior, or other concrete attack was established in the inspected runtime-operation locations.
Source & flagged code
7 flagged · loading sourcePackage source references child process execution.
dist/update/self-update.jsView on unpkg · L27Package source invokes a package manager install command at runtime.
dist/update/self-update.jsView on unpkg · L17Findings
Affected versions and remediation
This report applies to @telora/daemon@0.23.54.
See all versions of @telora/daemon for other recorded verdicts.
- Review the evidence and your use of @telora/daemon@0.23.54 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.