Oline

• Risk level: Low
• Analyzed: Mar 6, 2026, 5:06 AM
• Findings: 0
• Version: 1.0.0

Overview

This instruction-only coaching skill is internally consistent with its stated purpose: it stores a per-user profile and session state under a workspace, checks GitHub for updates, and uses that state to run Socratic dialogues and send nudges.

This skill appears to do what it says, but before installing consider: (1) It will create and update files under {workspace}/oline/ (profile.json, sessions.json, trajectory.md, .last-update-check). Review and secure that workspace or run the skill in an isolated workspace if you prefer. (2) It records internal-only signals (contradictions_flagged, engagement latency) that are never shown to the user — if you dislike hidden tracking, ask for a version that makes these signals explicit or delete the workspace files periodically. (3) The only external network call is a public GitHub releases check; verify the repository (benfoden/oline) if you want to confirm the update source. (4) No credentials are requested, and nothing is installed system-wide. (5) The agent can invoke this skill autonomously by default (platform normal), and the skill may send proactive nudges — confirm you accept that behavior and the expected nudge cadence. If any of these points are unacceptable, do not install or run it in a shared/backup-enabled workspace.