All projects on Base matching “agent safety”

AI agent action safety: one-cent deterministic preflight for every consequential agent command. Returns risk, matched rule ids, and an exact machine-readable escalation call: full DestructGuard evidence for medium risk or the complete Action Safety workflow for high and critical risk.

Home safety lookup for homeowner and safety agents. Returns CPSC hazard and recall data for home products, with fire and electrical risk context.

ERC-20 token safety pre-check from on-chain data only (Base, Ethereum, Arbitrum, Optimism, Polygon, BSC): name/symbol/decimals/supply, owner and whether renounced, EIP-1967/minimal proxy detection, and a bytecode scan (incl. implementation) for mint, blacklist, adjustable-fee, max-tx, trading-switch and pause functions, with flags and a risk level. Run before swapping into an unknown token. $0.005

Trace an HTTP redirect chain to its final destination without downloading page bodies. Returns every hop (URL + status), the final URL, and safety warnings — link-safety grounding for agents before they visit or cite a URL. Clean JSON.

Base ERC-20 check before an agent buys or holds a token. Input: one contract address. Output: allow, warn, or block. Block: not a contract, not an ERC-20, zero supply, or the owner or token contract holds at least 50 percent. Warn: 20 to 50 percent held, proxy (EIP-1967, EIP-1167, or implementation()), or bytecode contains mint or pause. A selector does not prove the public can call it. Concentrat

Headless scraper proxy for AI agents at $0.01 per accepted scrape. Fetch a public HTTPS URL and return clean LLM-ready Markdown. Agent-safety guarantee: output is validated as clean text before serving; binary or garbled content is rejected with no charge.

Headless scraper proxy for AI agents at $0.01 per accepted scrape. Fetch a public HTTPS URL and return clean LLM-ready Markdown. Agent-safety guarantee: output is validated as clean text before serving; binary or garbled content is rejected with no charge.

Resolve an HTTP or HTTPS URL through redirects and return the complete redirect chain, HTTP status for each hop, final URL, and redirect safety findings. Use when an agent needs to determine where a URL actually leads before visiting, citing, downloading, or trusting it.

Token safety check (rug check) for AI agents before buying, sending or accepting a crypto token on Base, Ethereum or Solana: give its address, get name, supply, live on-chain price and pool liquidity, plus contract controls (owner, upgradeable proxy, mint, pause, blacklist, fee setters) or Solana mint and freeze authority, with plain warnings. Facts read from the chain, not a rating.

Score a URL for phishing risk before an agent opens or trusts it. Keyless, deterministic heuristics - typosquat / homoglyph of a known brand, punycode / mixed-script hosts, credentials-in-URL, raw-IP hosts, suspicious TLDs, over-deep subdomains - plus a best-effort domain-age check (young domains are a top phishing signal). Returns a 0-100 risk score, a SAFE / SUSPICIOUS / DANGEROUS verdict, and l

AIﾉアカリ☆ Akari OS agent integration kit. The public philosophy corpus remains free; this paid machine-ready bundle adds a decision policy, safety boundary, integration recipes, validation checklist, content hash, and settlement receipt for agent workflows. Structured JSON, Base USDC, pay-per-request, no account.

Foraging guide — what wild edibles to forage now: edible wild plants, mushrooms, and berries by region and season, with safety-first ID guides and dangerous-lookalike warnings. Worldwide. For homesteading and forager agents.

Avoid unnecessary browser launches with a lightweight preflight for AI agents. ZeroWorker inspects public pages before browser automation and returns static forms, field constraints, CAPTCHA/SPA signals, DNS/public-network safety checks, and a 3-state routing verdict: http_candidate, browser_likely, or undetermined. Use it to choose direct HTTP or escalate to a browser. No API key required. $0.003

x402 seller + agent-readiness audit of one origin: live 402 challenge per paid route (up to 5), price vs catalog, HEAD/garbage-header safety, llms.txt, /.well-known/x402, openapi.json, dead doc links. Read-only.

Use when an agent is about to pay another agent and needs one finished answer first: should you pay this payee? PayCheck ($0.50) runs Flow's whole checker stack on the payee — payee verdict (kind, activity, contract safety), wallet approval-exposure audit, and recent USDC settlement history on Base — returning one grade (ALLOW / REVIEW / BLOCK, 0-100), a verify_status, all checks, the evidence, an

Pro Yield in one call: live stablecoin lending venue rates (safety tiers, blend target, sleeve rates, provenance) plus vault state on HyperEVM — share price, total assets, live caps — and a wallet's position when ?address= is supplied. Use when an agent wants the whole Pro Yield picture for one $0.010 call instead of two.

Use before an AI agent pays a vendor or counterparty. Returns an action-oriented allow/review/block evidence preflight: legal-entity resolution, potential sanctions signals, RDAP domain evidence and company-domain consistency. x402 payment safety decision support only; not legal or transaction approval.

Screen an EVM wallet or contract address BEFORE paying it. Deterministic, keyless compliance / safety gate: checks the address against the OFAC sanctioned crypto-address list and a curated taint set (mixer, hack, scam) and returns a single BLOCK / REVIEW / ALLOW recommendation plus the matched sources, so an AI agent can decide whether to send funds. No API keys, no LLM, one x402 USDC micropayment

Composite 0-100 buy-safety score for any token: combines contract security analysis (honeypot, taxes, ownership risks) with live market depth (liquidity, volume, activity) into a single verdict: avoid | high-risk | caution | ok | strong. One call answers 'should my agent touch this token?'

x402 v2 paid EVM token contract risk API on Base and BSC. For 0.01 USDC, returns EIP-1967 proxy detection, owner control and renouncement status, and block-explorer source-code verification. Built for AI agents, bots, and automated token screening.

Call this when an agent needs current significant global disaster events — e.g. before routing logistics or assessing exposure. Returns ALL matching normalized GDACS multi-hazard alerts (earthquake, cyclone, flood, volcano, drought, wildfire); one call returns the whole active set (~55-76 Orange/Red events/year). Refreshed daily 06:30 UTC (up to 24h old); not for life-safety use. Filters: source,

Inspect a Base token contract, public reputation, holder count, and DEX liquidity before trading.