Plugin

claude-anti-mem

Claude gets things wrong in patterns. It invents library methods by analogy. It wraps a one-liner in argparse and logging. It assumes Snowflake when you said "dbt". Each mistake is usually obvious in hindsight, but Claude repeats the shape of the mistake across sessions because it has no record of them. claude-anti-mem fixes that with two skills working as a pair. anti-mem-collector writes: when you correct Claude, it logs the failure pattern to ~/.claude-anti-mem/anti-mem.md with a concrete Check: for next time. anti-mem-checker reads: before Claude commits to a substantive response, it consults the rap sheet and revises any draft that matches a logged failure. Plain markdown, no database, no daemon. Commit the rap sheet to a repo to share it with your team.

• Published source
• Repository

Analysed

11 October 2026

Last observed

11 October 2026

What the code does

Static analysis found none of the capabilities we track, such as shell commands, network requests or file writes, in the latest version.

Having a capability is not wrongdoing; many tools need these to do their job. It tells you what to check before you install.

This is public metadata, not a safety certification or complete assessment. Static analysis cannot establish what a tool does at runtime.