memory-keeper
Manages project structure, auto-saves artifacts, and performs recovery protocols. Use this skill at the end of tasks or when encountering errors.
Installation
Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.
View source ↗GuidesSkill instructions
Memory Keeper Skill
Purpose
To ensure the Structural Memory of the agent is preserved, artifacts are backed up, and the agent can recover from context limits.
Trigger Conditions
- Task completion (finishing a feature).
- User mentions
/retroor "save memory". - Agent encounters "Agent terminated due to error" or "Context Limit" errors.
Workflow Rules
1. Structural Integrity Check
- Verify
agent/rules/exists. - Verify
docs/exists. - Verify
.memory/exists.-
Action: If missing, suggest running
init-projector creating them.
-
Action: If missing, suggest running
2. Auto-Save Protocol
When a major task is done:
-
Format Timestamp:
YYMMDD_HHMM. -
Target Directory:
.memory/[Timestamp]_[TaskName]/. -
Action: Copy key artifacts (
task.md,implementation_plan.md,spec.md) to this folder.- Note: Do not perform the copy yourself if it's too heavy. Just remind the user or create a script.
3. Recovery Protocol (Agent Terminated Handler)
If the Agent is slow, buggy, or hits token limits:
- Advise User: "I am hitting context limits. Recommendation: Factory Reset."
-
Command: Suggest running:
./antigravity_toolkit.sh full -
Post-Reset: Remind user to:
- "Import rules and workflows" immediately after reset.
Examples
Read the usage examples in the source repository.
View source ↗Source history
- Indexed on
- 2026-01-20
- Index updated
- 2026-01-26
- Branch
- main
Index and sync dates describe this directory. View the source repository for code history.
GitHub ↗chat Comments (0)
Sign in to join the discussion and leave a comment.
No comments yet. Be the first to share your thoughts!