Pattern Analyst

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill has no code or network behavior, but it asks the agent to persist and automatically update inferred personal profile patterns without explicit confirmation.

Install only if you want the agent to maintain a persistent pattern log and profile about you. Before using it, consider asking the agent to confirm every USER.md update first and periodically review notes/patterns.md and USER.md for inaccurate or overly sensitive inferences.

Findings (1)

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.

What this means

Incorrect or sensitive inferences about the user could be stored and reused in later conversations before the user has approved them.

Why it was flagged

This directs the agent to write inferred personal patterns into persistent profile memory without prior explicit user confirmation, which can influence future agent behavior.

Skill content
When a pattern repeats 3+ times across different interactions, auto-confirm it: ... Update `USER.md` immediately ... Hidden patterns ... are especially valuable — surface these even if auto-confirmed.
Recommendation

Require explicit confirmation before writing to USER.md, keep auto-confirmed observations in a separate reviewable log, and provide a clear way to delete or correct stored patterns.