Superlative Memory Manager
v1.0.0Unified memory management combining auto-compaction, tiered storage, semantic recall, and token optimization for efficient, lossless context handling in Open...
⭐ 0· 51·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name and description match the runtime instructions: the skill orchestrates existing memory-related skills (cognitive-memory, memory-tiering, context-compactor, memory-on-demand). No unrelated binaries or credentials are requested.
Instruction Scope
SKILL.md is instruction-only and stays within orchestration: enabling in openclaw.json, automatic operation, and optional CLI commands (memory compact, memory recall, memory store). However, it references WAL + git-backup, Git Notes, LanceDB/Lance vectors, and Qdrant without specifying how backups are performed, where data is written, or what git remotes/paths/credentials are used — this is a functional gap (not necessarily malicious) that affects data flows and privacy.
Install Mechanism
No install spec and no code files — instruction-only reduces installation risk. Nothing is downloaded or written by this skill itself.
Credentials
The skill requests no environment variables or credentials itself, which is coherent for a pure orchestrator. But because it claims to use external stores (vector DBs, git backups, WAL), expect that the underlying required skills will need credentials/configuration; the SKILL.md does not document or warn about where those credentials are configured or what the backup destination is.
Persistence & Privilege
always is false and the skill is user-invocable. It does not request persistent system-wide privileges or modify other skills' configs beyond the example of enabling itself in openclaw.json (normal behavior). Autonomous invocation is allowed by default (platform norm).
Assessment
This skill is primarily an orchestrator and appears coherent with its stated purpose, but confirm a few things before enabling:
- Verify the required underlying skills (cognitive-memory, memory-tiering, context-compactor, memory-on-demand) are trusted and inspect how they store data and manage credentials.
- Ask where 'git-backup' and WAL writes go and how git credentials/remote are provided — enabling this skill may cause automatic backups to a repository or filesystem you need to control.
- Check vector-store configuration (LanceDB/Qdrant): these stores often require endpoints/keys; ensure those credentials are scoped and stored safely by the required skills.
- Test in a non-production agent first to observe event emissions and storage/tiering behavior and ensure retention policies behave as expected.
If you need higher assurance, request the implementer to document exact backup destinations, credential requirements, and the CLI commands' provenance (which binary provides the memory CLI).Like a lobster shell, security has layers — review code before you run it.
latestvk97c575bm6e9z05998xewnctn183s2t2
License
MIT-0
Free to use, modify, and redistribute. No attribution required.