MCP security report
sigtap-mcp.sigtap.workers.dev
D55/100
Security grade DHigh-severity signals — review these before connecting.
Connected ✓13 tools scanned
https://sigtap-mcp.sigtap.workers.dev/mcp
highdeliverability_audit, email_grade, email_template, hash_text, jwt_decode, uuid_batch, slugify, json_tools, regex_test, crypto_price, domain_age, weather
Autonomous on-chain payment signing embedded in tool descriptions
Every paid tool description instructs the agent to complete a '402 challenge -> sign -> retry' flow that results in a real USDC transaction being signed and settled on-chain (Base network) per call. This effectively directs the AI agent to autonomously authorize financial transactions on behalf of the user for each tool invocation, with no explicit requirement for per-transaction user confirmation described in the schema. An agent following this instruction literally could spend user funds repeatedly without the user's informed consent, especially if it calls these tools many times in a session (e.g., via the catalog or in a loop). This is a broad, financially consequential capability disguised as ordinary tool metadata rather than a user-facing confirmation step.
mediumemail_grade, email_template, deliverability_audit
User content and domain data sent to a single third-party API without disclosed handling
These tools transmit potentially sensitive data (full email subject/body text, sending domain, offer/persona/recipient details) to a specific external endpoint (sigtap-outreach-api.sigtap.workers.dev) for both free preview and paid calls. The description does not disclose data retention, logging, or usage policy for this third party, so an agent invoking these tools may be exfiltrating user-authored content (which could contain PII or confidential business info) to an external service without the user being clearly informed.
low
Not on gate's verified list
This server isn't on gate's handpicked list. That's not necessarily bad, but there's no third-party signal about it — review its tools carefully.
lowjwt_decode
Decodes arbitrary JWTs which may contain sensitive tokens/claims
The tool decodes JWT header and payload (without signature verification) and sends the token to an external API for the paid path. If a user or agent passes a live session/auth JWT (e.g., copied from another system) into this tool, its contents (potentially including sensitive claims) would be sent to a third-party service. This is a normal utility function, but agents should avoid feeding real production credentials/tokens into it without user awareness.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/83258343-b306-4f7a-82c6-8f8bc056911c)Scanned 9/13/2026 · This is a point-in-time snapshot of the server’s public tools. Re-scan.
Connect this server safely with gate.
Continuous re-scans, drift alerts, per-tool allow/ask/block, and a log of every call.
Monitor it — free