Qmistake

To record AI session mistakes, flag the error or run the /Qmistake command. The system stores the entry in .qe/MISTAKE.md, auto-numbers it, and injects unresolved mistakes into future session context for cross-session awareness.

Cross-session mistake logging works by storing corrections in a centralized .qe/MISTAKE.md registry. At every new session start, the system loads and injects unresolved mistakes into the context to maintain continuous workflow awareness.

Yes, you can review and resolve logged errors. The system provides subcommands to list, review, and resolve mistake entries, allowing you to manage the .qe/MISTAKE.md registry and clear resolved issues from the active context.

The best way to maintain an AI error registry is to use a centralized logging file like .qe/MISTAKE.md. This approach auto-numbers entries and automatically loads unresolved issues at session start to ensure continuous awareness.

No dependencies or specific components are required. The mistake tracking system operates independently by writing to a local .qe/MISTAKE.md file and injecting those entries into the session context without external modules.

You should use it when you need automated cross-session awareness of errors. Instead of manual documentation, it detects flagged corrections, auto-numbers the entries, and actively injects unresolved mistakes into new sessions to prevent repeats.