Memory
Split memory system with USER.md for durable personal profile and MEMORY.md for token-budgeted operational context.
Install
$ agentstack add skill-rikyz90-shibaclaw-memory ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-rikyz90-shibaclaw-memory) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Memory
Structure
USER.md— Durable personal profile and preferences. Not token-compacted; keep it focused on long-lived user facts.memory/MEMORY.md— Operational long-term facts. Injected into every system prompt under# Memory, truncated from the bottom if it exceeds the token budget (~2000 tokens default).memory/HISTORY.md— Append-only log with[YYYY-MM-DD HH:MM] [#tag1 #tag2]entries. Never injected. Search it withmemory_searchor grep when historical context is missing.
MEMORY.md Layout
Sections are ordered by survival priority — top sections persist under truncation, bottom sections are dropped first.
## Environment— OS, runtime, tooling constraints, local services, provider setup## Entities— people, projects, repos, services referenced often## Project State— milestones, blockers, medium-term status, important decisions## Dynamic Context— current tasks, recent decisions, in-progress work
Rules:
- One fact per bullet, no prose paragraphs
- Update/replace existing facts instead of appending duplicates
- Put personal profile and preferences in
USER.md, not inmemory/MEMORY.md - Put durable operational facts in the top three MEMORY sections; only transient state goes in Dynamic Context
- Keep the file concise — the system auto-compacts when it exceeds ~1600 tokens
Missing Context
If a topic feels incomplete, search HISTORY.md first before assuming a fact was never recorded. Use the memory_search tool for semantic queries, or grep for exact matches:
grep -i "keyword" memory/HISTORY.mdProactive Context Retrieval
If the user refers to past discussions or ongoing workflows you don't recall: search HISTORY.md before responding.
Auto-consolidation
Handled automatically every ~10 messages. Long-term memory is auto-compacted by the system when it grows beyond the configured threshold.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: RikyZ90
- Source: RikyZ90/ShibaClaw
- License: Apache-2.0
- Homepage: https://deepwiki.com/RikyZ90/ShibaClaw
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.