KryptoGO Meme Trader

SuspiciousAudited by ClawScan on May 10, 2026.

Overview

This crypto trading skill has a clear purpose, but its monitor/cron path can automatically sell from a Solana wallet even though the docs say trades require confirmation by default.

Do not install or fund this skill with a main wallet. If you test it, use a dedicated low-value Solana wallet, avoid running the default or autonomous cron setup until the auto-sell behavior is fixed, check `openclaw cron list` for active jobs, and remember that the analysis backend is marked deprecated/shutting down.

Findings (6)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

A portfolio check or scheduled monitor can sell wallet holdings automatically, potentially causing financial loss or unwanted tax/trading consequences.

Why it was flagged

When a stop-loss, take-profit, or cluster-dump condition is detected, the monitoring script directly calls swap.py with --sell. The script does not prompt for user confirmation before attempting the sale.

Skill content
if action: ... cmd = ["python3", script_path, mint, str(balance), "--sell"] ... subprocess.run(cmd)
Recommendation

Make monitoring dry-run/notify-only by default, require an explicit --execute flag and user confirmation for supervised mode, and separate autonomous trading scripts from default monitoring.

What this means

Users may trust the default setup as advisory-only when it can actually trigger wallet transactions through the monitor script.

Why it was flagged

The default supervised cron setup says trades will not be auto-executed while instructing the agent to run monitor.py, which itself can execute sells.

Skill content
Trades will NOT be auto-executed... Run the monitoring script: 'source ~/.openclaw/workspace/.env && python3 skills/kryptogo-meme-trader/scripts/monitor.py'... Do NOT auto-execute sells.
Recommendation

Correct the documentation and default cron workflow so the default path cannot execute trades; clearly label any script that can trade and require explicit opt-in.

ConcernHigh Confidence
ASI10: Rogue Agents
What this means

A scheduled job could keep monitoring and trading after the initial interaction unless the user notices and removes it.

Why it was flagged

This creates a recurring background job that loads credentials and runs the monitor every five minutes. Because monitor.py can call swap.py to sell, the job can continue exercising wallet authority after setup.

Skill content
openclaw cron add --every 5m --name "stop-loss-tp" --message "Run the monitoring script: 'source ~/.openclaw/workspace/.env && python3 skills/kryptogo-meme-trader/scripts/monitor.py'..."
Recommendation

Require explicit consent before installing recurring trading jobs, show the user active cron jobs, and ensure teardown is easy and prominently documented.

What this means

If the .env file or sourced environment is exposed, the funded wallet could be drained.

Why it was flagged

The setup script creates or stores a Solana private key in the workspace .env file. This is expected for local signing, and the script sets chmod 600, but the credential can spend wallet funds.

Skill content
ENV_FILE = os.path.expanduser("~/.openclaw/workspace/.env") ... new_lines.append(f"SOLANA_PRIVATE_KEY={new_private_key}\n")
Recommendation

Use a dedicated low-value wallet, keep minimal funds in it, restrict access to the .env file, and avoid sharing the workspace with untrusted skills or users.

What this means

Future dependency changes or compromised packages could affect the trading environment.

Why it was flagged

The setup script installs runtime dependencies by package name without pinned versions or hashes. This is purpose-aligned but leaves dependency provenance/version risk.

Skill content
REQUIRED_PACKAGES = ["solders", "requests"] ... [sys.executable, "-m", "pip", "install", pkg]
Recommendation

Pin dependency versions, prefer a virtual environment, and review packages before running setup.

What this means

Incorrect or tampered memory files could change trading thresholds or future recommendations/actions.

Why it was flagged

The cron workflow uses persistent memory files for preferences, lessons, state, and trade journals that can influence later trading decisions.

Skill content
Read memory/trading-preferences.json... Read memory/trading-lessons.md... Update memory/trading-state.json... trading-journal.json
Recommendation

Review and protect the memory files, reset them if behavior seems wrong, and avoid letting untrusted content modify trading preferences or lessons.