mirror of
https://github.com/cursor/plugins.git
synced 2026-09-14 20:00:00 +08:00
2cf89ae30b
Rename the learning plugin slug to teaching across marketplace metadata and docs, and add the reused data-science avatar to continuous-learning while preserving plugin validation. Co-authored-by: Cursor <cursoragent@cursor.com>
Continuous Learning
Automatically and incrementally keeps AGENTS.md up to date from transcript changes.
The plugin combines:
- A
stophook that decides when to trigger learning. - A
continuous-learningskill that mines only high-signal transcript deltas.
It is designed to avoid noisy rewrites by:
- Reading existing
AGENTS.mdfirst and updating matching bullets in place. - Processing only new or changed transcript files.
- Writing plain bullet points only (no evidence/confidence metadata).
Installation
/add-plugin continuous-learning
How it works
On eligible stop events, the hook may emit a followup_message that asks the agent to run the continuous-learning skill.
The hook keeps local runtime state in:
.cursor/hooks/state/continuous-learning.json(cadence state)
The skill uses an incremental transcript index at:
.cursor/hooks/state/continuous-learning-index.json
Trigger cadence
Default cadence:
- minimum 10 completed turns
- minimum 120 minutes since the last run
- transcript mtime must advance since the previous run
Trial mode defaults (enabled in this plugin hook config):
- minimum 3 completed turns
- minimum 15 minutes
- automatically expires after 24 hours, then falls back to default cadence
Optional env overrides
CONTINUOUS_LEARNING_MIN_TURNSCONTINUOUS_LEARNING_MIN_MINUTESCONTINUOUS_LEARNING_TRIAL_MODECONTINUOUS_LEARNING_TRIAL_MIN_TURNSCONTINUOUS_LEARNING_TRIAL_MIN_MINUTESCONTINUOUS_LEARNING_TRIAL_DURATION_MINUTES
Output format in AGENTS.md
The skill writes only:
## Learned User Preferences## Learned Workspace Facts
Each item is a plain bullet point.
License
MIT