aawp-ai@2.0.0
CAUTION· install onlyPublisher claims: "AAWP (AI Agent Wallet Protocol) — the only crypto wallet protocol built exclusively for AI Agents on EVM-compatible blockchains and Solana. Not for humans. The signer is the AI Agent itself, cryptographically bound at wallet creation. Supports wallet lifecycle management, token transfers, DEX swaps, cross-chain bridging, Solana native trading (Pump.fun SDK + Jupiter), token launches, DCA automation, and price alerts." Observed in sandbox: calls raw.githubusercontent.com only, touches nothing outside the project, writes no memory.
Coverage
Install scripts ran; no agent task scenario for this bundle.- Tested
- install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; install scripts ran; no agent task scenario for this bundle) on 2026-11-07 and scored it 79/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; Destructive file system command detected; Financial transfer language. Summary: calls raw.githubusercontent.com only, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenariosStatic findings
try { const p = require('path'); require('fs').readFileSync(p.join(__dirname, '..', '.env'), 'utf8').split('\n').forEach(l => { const [k,...v] = l.split('='); if (k && !process.env[k.trim()]) process.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.# Only kill if it's actually running from the aawp skill dirFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.kill "$pid" 2>/dev/nullFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.rm -f /tmp/aawp.log /tmp/aawp-start.out /tmp/aawp-ensure.out # clear old logsFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.The only crypto wallet protocol built exclusively for AI Agents. Not for humans.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.if ! kill -0 "$DAEMON_PID" 2>/dev/null; thenFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.console.log(`Sending via AAWP wallet ${walletAddr}...`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.// Load .env for AAWP_WALLET pin and other configFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.chmod +x "$NODE_FILE" 2>/dev/null || trueFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.AAWP (AI Agent Wallet Protocol) — the only crypto wallet protocol built exclusivelyFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.kill "$pid" 2>/dev/null || trueFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.console.log('Usage: wallet-manager [--chain <chain>] send-token <TOKEN|0xAddr> <to> <amount>');Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(' ✅ Signature posted');Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(`Sending ${amount} ${chain.nativeCurrency} on ${chain.name}`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.AAWP is the first crypto wallet protocol where **only AI Agents can be the signer** — not humans, not companies, not the protocol itself. This is enforced at the contract level, not a policy.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.if ! kill -0 "$DAEMON_PID" 2>/dev/null; thenFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.The only crypto wallet protocol built exclusively for AI Agents.<br>Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.console.log(`Sending: ${RAW_AMOUNT.toString()} to ${TO}`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(` You send: ${amtIn} ${symIn} (${fromChain.toUpperCase()})`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(`Sending ${amount} ${sym} on ${chain.name}`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log(` Need: ${ethers.formatEther(sendAmt + MIN_WALLET_BAL)} (incl. reserve)`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.console.log('Usage: wallet-manager [--chain <chain>] send <to> <amount>');Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
1 point to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim wfce: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- File type combined with upload action≈ +2 on the totalRemove the call, or declare the host in frontmatter and explain in the README what is sent and why.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.