Skill flagged β€” suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Memphis Cognitive

🧠 Memphis Cognitive Engine - Complete AI Memory System Transform your OpenClaw agent into a cognitive partner with: - Model A: Record conscious decisions (m...

MIT-0 Β· Free to use, modify, and redistribute. No attribution required.
⭐ 0 · 276 · 2 current installs · 2 all-time installs
MIT-0
Security Scan
VirusTotalVirusTotal
Suspicious
View report β†’
OpenClawOpenClaw
Suspicious
medium confidence
β„Ή
Purpose & Capability
The skill is a META-PACKAGE whose SKILL.md and wrapper script simply delegate to an external Memphis CLI; requiring the external CLI is appropriate for the stated purpose. However, registry metadata (v3.7.2) and included _meta.json / README versions (3.6.7 / 3.6.3) disagree, which is an inconsistency worth questioning.
β„Ή
Instruction Scope
SKILL.md stays within the expected scope: it only instructs installing and using the Memphis CLI and related optional tools (Ollama, Node.js). It references local paths like ~/memphis and ~/.memphis which are expected for a CLI. It does recommend networked features (multi-agent sync, trade manifests) β€” expected for this product but these features imply network communication and key/identity use that the user should understand before enabling.
!
Install Mechanism
There is no built-in install spec; the instructions ask you to run curl -fsSL https://raw.githubusercontent.com/.../install.sh | bash and other remote install scripts. Downloading and executing a remote install script directly is a common but higher-risk pattern. The URLs are GitHub raw and referenced project domains (not obscure shorteners), which reduces but does not eliminate risk. No other suspicious download hosts were used.
βœ“
Credentials
The skill declares no required environment variables or secrets and the wrapper script simply forwards arguments to the memphis CLI. The SKILL.md mentions local config files (~/.memphis) and data directories, which is expected. No unrelated credentials or broad environment access are requested.
βœ“
Persistence & Privilege
always:false and the skill is instruction-only with a thin wrapper script that does not modify other skills or global agent settings. It does not request persistent elevated privileges in the skill bundle itself.
What to consider before installing
This skill is mostly documentation that expects you to install the Memphis CLI from the author's GitHub. Before installing or piping any remote script to bash: 1) Inspect the referenced install.sh on GitHub (do not run it blind). 2) Prefer cloning the repo and building locally (git clone + npm install + npm run build) if you can. 3) Verify the GitHub repo/owner legitimacy (stars, recent commits, tags, signed releases). 4) Note the mismatched version strings in the skill metadata β€” ask the publisher why versions differ. 5) Be aware that multi-agent sync and trade features perform network operations and use cryptographic identities (DIDs); review what data will be shared and where. 6) If unsure, run the installer in an isolated environment (container or VM) and examine ~/.memphis and any generated keys before connecting to other agents. If you want, I can: fetch and summarize the install.sh contents, compare repository tags/releases, or highlight specific lines in install scripts to review.

Like a lobster shell, security has layers β€” review code before you run it.

Current versionv3.7.2
Download zip
aivk9707dhadbmeydqefjenm78ryx824g88cognitivevk9707dhadbmeydqefjenm78ryx824g88decisionsvk9707dhadbmeydqefjenm78ryx824g88latestvk9760fe9xfxawgbr230hccahwh82aq3blocal-firstvk9707dhadbmeydqefjenm78ryx824g88memoryvk9707dhadbmeydqefjenm78ryx824g88polskavk9707dhadbmeydqefjenm78ryx824g88

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

SKILL.md

Memphis Cognitive Engine v3.7.2

πŸ“’ IMPORTANT: This is the MAIN Memphis skill

βœ… ONLY MEMPHIS-COGNITIVE IS ACTIVELY MAINTAINED

Other Memphis skills on ClawHub (memphis, memphis-bootstrap, memphis-brain, memphis-super) are DEPRECATED.
Please use this skill only: clawhub install memphis-cognitive


πŸŽ‰ PRODUCTION READY - 100% WORKING, ZERO BUGS!

Status (2026-03-05 02:03 CET):

  • βœ… 17/17 commands working (100%)
  • βœ… Zero critical bugs
  • βœ… All advanced features tested
  • βœ… Multi-agent network operational
  • βœ… Binary builds available (Linux, macOS, Windows)
  • βœ… Consolidated to single skill (2026-03-05)

⚑ Installation (5 minutes)

⚠️ IMPORTANT: This skill is a META-PACKAGE. Memphis CLI must be installed separately.

Step 1: Install Memphis CLI (4 min)

# Option 1: One-liner (RECOMMENDED)
curl -fsSL https://raw.githubusercontent.com/elathoxu-crypto/memphis/main/install.sh | bash

# Option 2: Manual
git clone https://github.com/elathoxu-crypto/memphis.git ~/memphis
cd ~/memphis
npm install && npm run build
npm link  # Or: npm run install-global

# Option 3: From npm (when published)
npm install -g @elathoxu-crypto/memphis

Step 2: Install ClawHub Skill (1 min)

clawhub install memphis-cognitive

Step 3: Initialize (30 sec)

memphis init  # Interactive setup wizard

βœ… Done! Memphis is ready! πŸŽ‰


πŸš€ Quick Start (5 minutes)

1. First Decision (30 sec)

memphis decide "Use TypeScript" "TypeScript" -r "Better type safety"

2. Ask Memory (30 sec)

memphis ask "Why did I choose TypeScript?"

3. Search Memory (30 sec)

memphis recall "typescript"

4. Advanced: Knowledge Graph (30 sec)

memphis graph build --limit 50
memphis graph show --stats

5. Advanced: Reflection (30 sec)

memphis reflect --daily

βœ… Memory system operational!


🧠 The 3 Cognitive Models

Model A: Conscious Decisions (Manual)

Speed: 92ms average (frictionless capture)

# Full syntax
memphis decide "Database choice" "PostgreSQL" \
  -r "Better JSON support" \
  -t tech,backend \
  --scope project

# Frictionless (92ms)
memphis decide-fast "use TypeScript"

Model B: Inferred Decisions (Automatic)

Accuracy: 50-83% confidence

# Analyze last 30 days
memphis infer --since 30

# Interactive mode
memphis infer --prompt --since 7

Model C: Predictive Decisions (Predictive)

Accuracy: 78% average (improves over time)

# Learn patterns (one-time setup, needs 50+ decisions)
memphis predict --learn --since 90

# Get predictions
memphis predict

# Proactive suggestions
memphis suggest --force

# Track accuracy
memphis accuracy

πŸš€ Advanced Features (v3.6.2)

1. TUI Dashboard

Interactive terminal UI for monitoring

memphis tui

Features:

  • Real-time chain monitoring
  • Visual block explorer
  • Dashboard statistics

2. Knowledge Graph

Visualize knowledge connections

# Build graph
memphis graph build --limit 50

# Show statistics
memphis graph show --stats

# Filter by chain
memphis graph show --chain journal --depth 2

Performance: 50 nodes, 1778 edges in 36ms βœ…


3. Reflection Engine

Generate insights from memory

# Daily reflection (last 24h)
memphis reflect --daily

# Weekly reflection (last 7 days)
memphis reflect --weekly

# Deep dive (last 30 days)
memphis reflect --deep

# Save to journal
memphis reflect --weekly --save

Features:

  • Automatic pattern detection
  • Curiosity tracking
  • Recurring concept identification
  • Knowledge graph clusters

Performance: 46ms for 179 entries βœ…


4. Trade Protocol

Multi-agent knowledge exchange

# Create trade manifest
memphis trade create "did:memphis:recipient-123" \
  --blocks journal:100-110 \
  --ttl 7 \
  --out /tmp/trade.json

# Verify signature
memphis trade verify /tmp/trade.json

# Accept trade
memphis trade accept /tmp/trade.json

# List pending offers
memphis trade list

Features:

  • DID-based identity
  • Cryptographic signatures
  • TTL (time-to-live) support
  • Usage rights specification

5. Multi-Agent Sync

Synchronize between Memphis instances

# Check sync status
memphis share-sync --status

# Push local blocks to remote
memphis share-sync --push

# Pull remote blocks to local
memphis share-sync --pull

# Custom remote
memphis share-sync --push --remote 10.0.0.80 --user memphis

Performance: 18 blocks synced in 0.81s βœ…


6. Decision Management

List and manage decisions

# List all decisions
memphis decisions list

# Filter by tags
memphis decisions list --tags project,tech

# Limit results
memphis decisions list --limit 5

# JSON output
memphis decisions list --json

# Show by numeric index
memphis show decision 0
memphis show decision 1

# Show by decisionId
memphis show decision e4b5877bf16abae5

πŸ“š Complete Command Reference

Core Commands (17 total - 100% working)

Decisions:

  • memphis decide - Record decision (Model A)
  • memphis decide-fast - Ultra-fast capture (<100ms)
  • memphis decisions list - List all decisions
  • memphis show decision <id> - Show decision details
  • memphis revise - Revise decision
  • memphis infer - Detect from git (Model B)
  • memphis predict - Predict decisions (Model C)
  • memphis suggest - Proactive suggestions

Memory:

  • memphis journal - Add journal entry
  • memphis ask - Query with context
  • memphis recall - Semantic search
  • memphis reflect - Generate insights

Advanced:

  • memphis tui - Terminal UI dashboard
  • memphis graph build - Knowledge graph
  • memphis trade create/accept - Trade protocol
  • memphis share-sync - Multi-agent sync
  • memphis status - System status

🎯 Real-World Examples

Scenario 1: Developer Starting New Project

# 1. Architecture decisions (Model A)
memphis decide "API Architecture" "GraphQL" \
  -r "Flexible queries for frontend"
  
memphis decide-fast "use PostgreSQL for main DB"

# 2. Work on project, commit to git
git commit -m "Migrated from REST to GraphQL"

# 3. Detect decisions from commits (Model B)
memphis infer --since 7

# 4. Build knowledge graph
memphis graph build --limit 50
memphis graph show --stats

# 5. Daily reflection
memphis reflect --daily --save

Scenario 2: Multi-Agent Coordination

# On Agent 1 (Watra - Testing)
memphis decide "Network topology" "Watra (testing) ↔ Memphis (production)"

# Create trade offer
memphis trade create "did:memphis:memphis-production" \
  --blocks journal:100-110 \
  --ttl 7 \
  --out /tmp/trade.json

# Sync share chain
memphis share-sync --push

# On Agent 2 (Memphis - Production)
memphis trade accept /tmp/trade.json
memphis share-sync --pull

Scenario 3: Weekly Review

# 1. Generate weekly reflection
memphis reflect --weekly --save

# 2. Build knowledge graph
memphis graph build --limit 100

# 3. Check decision patterns
memphis predict --learn --since 90
memphis predict

# 4. Sync with team
memphis share-sync --push

πŸ“Š Performance

FeatureTest SizeTimeTargetStatus
Decision Capture1 decision92ms<100msβœ… 8% faster
Embedding5 blocks159ms<500msβœ… 68% faster
Graph Build50 nodes36ms<50msβœ… 28% faster
Reflection (daily)179 entries46ms<100msβœ… 54% faster
Trade Create3 blocks<1s<2sβœ… Instant
Multi-Agent Sync18 blocks0.81s<5sβœ… 84% faster

Overall: All targets exceeded! βœ…


πŸ”§ Requirements

Minimum:

  • Node.js 18+
  • Git 2.x+
  • 500MB disk space

Recommended:

  • Node.js 20+
  • Ollama (for embeddings)
  • 2GB RAM
  • SSH access (for multi-agent sync)

Install Ollama:

curl -fsSL https://ollama.com/install.sh | sh
ollama pull nomic-embed-text

πŸš€ What's New in v3.6.2

New Features:

  • βœ… TUI Dashboard - Interactive terminal UI (fixed)
  • βœ… Knowledge Graph - Visualize connections (36ms, 1778 edges)
  • βœ… Reflection Engine - Generate insights (46ms)
  • βœ… Trade Protocol - Multi-agent knowledge exchange
  • βœ… Share-Sync - Multi-agent synchronization (0.81s for 18 blocks)
  • βœ… Decisions List - List all decisions with filtering
  • βœ… Numeric Index - memphis show decision 0 now works

Bug Fixes:

  • βœ… TUI blessed import fixed (Commit 8bcc930)
  • βœ… Onboarding config bugs fixed (Commit ba4bf7b)
  • βœ… Genesis blocks auto-creation (Commit ba4bf7b)
  • βœ… Decision ID confusion fixed (Commit 24a26a6)

Status:

  • βœ… 100% working (17/17 commands)
  • βœ… Zero critical bugs
  • βœ… Production ready
  • βœ… Multi-agent network operational

πŸŽ“ Learning Path

Beginner (30 min)

  1. Install Memphis CLI + skill (10 min)
  2. Make first decision (5 min)
  3. Search memory (5 min)
  4. Try examples (10 min)

Intermediate (1 hour)

  1. Use all 3 cognitive models (30 min)
  2. Try advanced features (20 min)
  3. Review documentation (10 min)

Advanced (2 hours)

  1. Pattern learning deep dive (30 min)
  2. Multi-agent setup (30 min)
  3. Knowledge graph exploration (30 min)
  4. Custom integration (30 min)

πŸ”’ Privacy & Security

  • βœ… Local-first (data stays on your machine)
  • βœ… No telemetry or tracking
  • βœ… No cloud required
  • βœ… Open source (MIT license)
  • βœ… End-to-end encryption (sync)
  • βœ… You own your data

🀝 Support

Get Help:

Contribute:


πŸŽ‰ Production Ready!

Memphis v3.6.2 is production-ready with:

  • βœ… 100% working commands (17/17)
  • βœ… Zero critical bugs
  • βœ… Multi-agent network operational
  • βœ… First real sync complete
  • βœ… All advanced features tested
  • βœ… Active development and support

Start your cognitive journey today!

git clone https://github.com/elathoxu-crypto/memphis.git ~/memphis
cd ~/memphis && npm install && npm run build && npm link
clawhub install memphis-cognitive
memphis status

Made with 🧠 by Elathoxu Abbylan (Memphis)

Repository: https://github.com/elathoxu-crypto/memphis
ClawHub: https://clawhub.com/skill/memphis-cognitive
Version: 3.6.2
Status: βœ… Production Ready (100% Working, Zero Bugs)
Last Updated: 2026-03-04 19:42 CET


Memphis: Record β†’ Detect β†’ Predict β†’ Sync πŸš€

Files

5 total
Select a file
Select a file to preview.

Comments

Loading comments…