OmniFocus
Analysis
This skill appears to do what it says—manage OmniFocus tasks locally—but users should confirm before letting it change tasks.
Findings (2)
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
if (allTasks[i].name() === taskName && !allTasks[i].completed()) { ... allTasks[i].markComplete(); ... return true; }The script can mutate persistent OmniFocus data by completing the first matching task name; this is purpose-aligned, and SKILL.md instructs confirmation before modifying tasks.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
const note = (task.note() || "").toLowerCase(); ... note: task.note() || ""
Search and list operations include task notes in returned JSON, so private OmniFocus task content can be brought into the agent conversation.
