remember

remember is a Memory stores & recall skill for AI agents, published by SparkDevNetwork in the Rock repository on GitHub. Skillspace rates remember D (62/100) and it passed the safety scan; the rating is based on static analysis only.

Audit the current session for things worth saving to auto-memory and present them as a plan-mode draft for explicit user approval before writing. Also identifies bubble-up candidates — rules that arguably belong in team-shared memory (CLAUDE.md or .claude/rules/) instead of private auto-memory — and surfaces them as suggestions without writing them. Use when the user says "remember", "/remember", "save what we learned", "capture this session", or wants a deliberate retrospective pass on a long session before context fades. Reads MEMORY.md plus every individual memory file first to dedupe; classifies each candidate as user / feedback / project / reference; drafts full bodies; awaits ExitPlanMode approval before writing anything. Do NOT use for: editing CLAUDE.md (that is the /memory slash command's job), saving session-specific task state (memory is for cross-session learning), or recording git history or code patterns that are derivable from the repo.

Grade

D

Quality

62 / 100

Safety

Passed

GitHub stars

670

Unproven: rated by static analysis only, not yet compared head-to-head.

This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.

Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.