Skip to main content

sweetai-agent-firewall

Pre-transaction trust firewall for AI agents. Ask "should I pay agent X $N?" before any agent-to-agent payment — one call, $0.01, no API key, no subscription (x402 / USDC).

from sweetai_firewall import SweetAIFirewall

fw = SweetAIFirewall()
result = fw.should_i_pay(agent_id=55985, amount_usd=25)
if result.payment_required:
    # Settle result.raw['accepts'] with your x402 wallet, then repeat the call.
    raise RuntimeError("Settle SweetAI x402 payment before continuing")
if result.action != "PAY":
    raise RuntimeError(f"Transfer blocked: {result.action}")
print(result.verdict)
# pay 0.01 USDC on eip155:8453, ... to 0x... to unlock the 25.0 USD decision

After your agent settles the 0.01 USDC payment (x402 X-PAYMENT header), the same call returns the full decision:

result.action          # "PAY" / "CAUTION" / "SKIP"
result.confidence      # 0.0 - 1.0
result.expected_loss_usd  # expected loss sized to YOUR amount

What's under the hood

  • ERC-8004 / Solana-8004 on-chain feedback analytics (Base, Ethereum, X Layer, Solana, Bittensor)
  • Sybil forensics: reviewer concentration (Herfindahl), ghost-reviewer ratio, burst detection
  • Liveness SLA: p50/p95 latency + flap counts, probed every 15 min
  • x402 pay-per-call: 0.01 USDC per decision on Base/Polygon/Arbitrum/Avalanche/Sei/X Layer/Solana

Free surfaces (no payment needed)

fw.status()               # live index stats
fw.network_sybil_overview()  # risk tiers across all indexed agents
fw.pricing()              # current prices

MCP alternative

All tools are also available as an MCP server for Claude Desktop / Cursor / any MCP agent:

{"mcpServers": {"sweetai-agent-intel": {"url": "https://sweetai.tiamatcrypto.com/mcp", "transport": "http"}}}

Docs: https://sweetai.tiamatcrypto.com/llms.txt · Manifest: /.well-known/x402 · agent.json: /agent.json

Metadata

Release files for sweetai-agent-firewall 0.1.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for sweetai-agent-firewall 0.1.3
File Size Uploaded Provenance
sweetai_agent_firewall-0.1.3.tar.gz 5.7 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for sweetai-agent-firewall 0.1.3
File Interpreter ABI Platform
sweetai_agent_firewall-0.1.3-py3-none-any.whl Python 3 none any Details

Total release size: 12.2 kB

Release files / sweetai_agent_firewall-0.1.3.tar.gz

Download URL sweetai_agent_firewall-0.1.3.tar.gz
Size 5.7 kB
Tags Source
SHA-256 checksum
How to use checksums
64736739f74b2663a8afc6fca3be3315cd84260f36ae777193a31238f7722523
BLAKE2b-256 checksum
How to use checksums
81ffadb47a0ce820691862d9df7a88b935456720c6966e73fa8e6f11f9a327be
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.11.15

Release files / sweetai_agent_firewall-0.1.3-py3-none-any.whl

Download URL sweetai_agent_firewall-0.1.3-py3-none-any.whl
Size 6.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
5d06381dd17e349fbf85be29be98324b563459e58ba94cea1cbb292974c9d818
BLAKE2b-256 checksum
How to use checksums
56ab09f12f11d5ba902cf4d2c239677cc7fa34e045c12e77d0ef96e2eea48ebb
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.11.15

Security for sweetai-agent-firewall 0.1.3

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page