中文 EN

Scan Report

Scan New Files

✓ Trusted — Risk Score 5/100
Last scan:17 hr ago Rescan
5 /100
elite-longterm-memory
Ultimate AI agent memory system combining WAL protocol, vector search, git-notes, and optional cloud backup
A straightforward AI agent memory management utility with no malicious indicators. All operations (file creation, optional network sync) are documented and benign.
Skill Nameelite-longterm-memory
Duration42.5s
Enginepi
✓
Safe to install
Approve for use. No security concerns identified.

Findings 2 items

Severity Finding Location
Low
SKILL.md declares 'Read' but skill performs file writes Doc Mismatch
The _meta.json allowed-tools section maps to filesystem:READ, yet the skill creates SESSION-STATE.md, MEMORY.md, memory/ directory, and daily logs — all WRITE operations. This is a minor doc/code mismatch for a benign file-writing tool.
allowed-tools mapping declares Read→filesystem:READ
→ Update _meta.json allowed-tools to include Write if the skill is expected to create files. This is a documentation hygiene issue, not a security concern.
 SKILL.md:1
Low
Shell command uses bare tilde without quotes in documentation Sensitive Access
The maintenance command `rm -rf ~/.openclaw/memory/lancedb/` uses an unquoted tilde (~). While this targets a specific subdirectory path and is clearly labeled as a 'nuclear option' cleanup, bare tilde expansion can be surprising. The command is fully documented and serves a legitimate cleanup purpose.
rm -rf ~/.openclaw/memory/lancedb/
→ Quote the path: `rm -rf "$HOME/.openclaw/memory/lancedb/"` to avoid any shell expansion surprises. This is a documentation best-practice, not an actual vulnerability.
 SKILL.md:293
ResourceDeclaredInferredStatusEvidence
Filesystem READ WRITE ✓ Aligned SKILL.md metadata declares 'Read' but the skill creates SESSION-STATE.md, MEMORY…
Network NONE NONE — SuperMemory/Mem0 cloud sync is described as OPTIONAL and requires explicit user …
Shell NONE NONE — No subprocess or shell execution in bin/elite-memory.js; rm -rf in SKILL.md is d…
Environment NONE NONE — Script only reads HOME env var for path construction; no iteration over env vars…
Clipboard NONE NONE — No clipboard access
Browser NONE NONE — No browser automation
Database NONE NONE — LanceDB is an optional plugin described in docs; no active database operations i…
Skill Invoke NONE NONE — No cross-skill invocation
1 Critical 17 findings
💀
Critical Dangerous Command 危险 Shell 命令
rm -rf ~
SKILL.md:293
🔗
Medium External URL 外部 URL
https://img.shields.io/npm/v/elite-longterm-memory.svg?style=flat-square
README.md:5
🔗
Medium External URL 外部 URL
https://www.npmjs.com/package/elite-longterm-memory
README.md:5
🔗
Medium External URL 外部 URL
https://img.shields.io/npm/dm/elite-longterm-memory.svg?style=flat-square
README.md:6
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/License-MIT-yellow.svg?style=flat-square
README.md:7
🔗
Medium External URL 外部 URL
https://opensource.org/licenses/MIT
README.md:7
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/Claude-AI-orange?style=for-the-badge&logo=anthropic
README.md:14
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/GPT-OpenAI-412991?style=for-the-badge&logo=openai
README.md:15
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/Cursor-IDE-000000?style=for-the-badge
README.md:16
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/LangChain-Framework-1C3C3C?style=for-the-badge
README.md:17
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/elite-longterm-memory
README.md:157
🔗
Medium External URL 外部 URL
https://x.com/NextXFrontier
README.md:162
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/bulletproof-memory
SKILL.md:401
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/lancedb-memory
SKILL.md:402
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/git-notes-memory
SKILL.md:403
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/memory-hygiene
SKILL.md:404
🔗
Medium External URL 外部 URL
https://clawdhub.com/skills/supermemory
SKILL.md:405

File Tree

5 files · 24.0 KB · 820 lines
Markdown 2f · 571L JavaScript 1f · 187L JSON 2f · 62L
├─ ▾ 📁 bin
│ └─ 📜 elite-memory.js JavaScript 187L · 4.8 KB
├─ 📋 _meta.json JSON 5L · 140 B
├─ 📋 package.json JSON 57L · 1.3 KB
├─ 📝 README.md Markdown 162L · 5.5 KB
└─ 📝 SKILL.md Markdown 409L · 12.4 KB

Dependencies 1 items

PackageVersionSourceKnown VulnsNotes
mem0ai ^1.0.0 npm (optionalDependencies) No Optional dependency, not installed by default, requires user-supplied API key

Security Positives

✓ No credential harvesting — skill does not read ~/.ssh, ~/.aws, .env, or iterate over process.env for secrets
✓ No network exfiltration — no outbound connections, no HTTP requests to external IPs
✓ No obfuscation — all code is plain JavaScript/Node.js, no base64, no eval()
✓ No supply chain risk — only optionalDependencies (mem0ai) is pinned to ^1.0.0, and it is optional and user-installed
✓ No persistence mechanisms — no cron jobs, no startup hooks, no backdoors
✓ No prompt injection — no hidden instructions in comments or strings
✓ No remote code execution — bin/elite-memory.js uses only fs/path built-ins
✓ File operations are scoped to the current working directory and explicitly documented paths under $HOME/.clawdbot
✓ Cloud sync features (SuperMemory, Mem0) are clearly optional and require explicit user API key setup