Never Twice tool released: letting AI agents check lessons before acting
Understand this story
Developer kaushaltrii88 released Never Twice, a tool that lets AI agents record and reuse lessons learned. Before performing change operations such as installing packages, modifying code, running migrations, or submitting deployments, agents first run `cognicore lesson check` to match relevant lessons and follow them, reducing repeat mistakes. Lessons are by default stored at project scope in the local CogniCore data directory (~/.cognicore/mcp/lessons.db), and the publisher says data does not leave the local machine; only lessons that apply to all projects get --scope global. The publisher also warns against saving passwords, tokens, keys, or personal data in lesson records.
AI overview based on source reports · Updated 1 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AINever Twice dont let the AI agent do mistake again
This guide describes a "Never Twice" workflow in which an AI agent stores lessons via the cognicore command and checks them before any task that changes something, such as installing packages, editing code, running tests, committing, pushing or deploying. Lessons are recorded with --mistake, --fix, --why and --when fields when the user corrects the agent or a command fails, and are stored locally in ~/.cognicore/mcp/lessons.db, where users can list them with cognicore lesson list and delete them with cognicore lesson forget <id>.
Activity for this story
- Current comparable activity
- 5
- Peak comparable activity
- 10Nov 13 22:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.