Skip to content

feat: add contract-risk-scanner skill#289

Open
believefnd wants to merge 2 commits intoBankrBot:mainfrom
believefnd:feat/contract-risk-scanner
Open

feat: add contract-risk-scanner skill#289
believefnd wants to merge 2 commits intoBankrBot:mainfrom
believefnd:feat/contract-risk-scanner

Conversation

@believefnd
Copy link
Copy Markdown

Summary

This PR adds a new skill: contract-risk-scanner.

The skill helps agents run pre-trade contract safety analysis before buy/swap execution, with a structured output:

  • Risk score (0-100)
  • Severity (low / medium / high / critical)
  • Confidence level
  • Top findings
  • Recommendation (GO / CAUTION / AVOID)

Why This Skill

Current ecosystem has strong coverage for execution, trading, and infra access.
What is still underrepresented is a dedicated pre-trade safety gate focused on:

  • Honeypot / transfer traps
  • Owner / privileged controls
  • Upgradeability risk
  • Blacklist / whitelist restrictions
  • Fee / tax and sellability risk

This skill is designed to reduce avoidable losses from unsafe contracts.

What's Included

  • New skill definition for contract-risk-scanner
  • Trigger-focused description for safety and pre-trade intent
  • Usage examples and output schema
  • Safety rules and uncertainty handling
  • Error handling and recommended agent workflow

Example User Intents Covered

  • Is this token safe?
  • Check if this CA is honeypot
  • Can I buy and sell this token safely on Base?
  • Run pre-trade risk report before swapping 100 USDC

Testing Performed

  • Verified markdown / frontmatter structure
  • Verified trigger phrasing for safety-related intents
  • Verified output format consistency (score / severity / findings / recommendation)
  • Verified explicit uncertainty handling for incomplete or unverified contracts

Notes

This skill does not claim absolute safety and always includes confidence limits.
For high-risk or incomplete data, it forces CAUTION or AVOID behavior.

Bankr Skill Contributor and others added 2 commits March 31, 2026 12:45
…rences

- Rewrote SKILL.md with GoPlus Security API + DexScreener as concrete data sources
- Added real curl examples, response field mappings, and risk scoring matrix
- Added Bankr integration gating pattern (GO/CAUTION/AVOID → bankr exec)
- Added scripts/scan.sh: full risk scan (honeypot + ownership + liquidity)
- Added scripts/honeypot.sh: quick honeypot-only check
- Added scripts/liquidity.sh: LP status via DexScreener
- Added references/goplus-api.md: full field reference + chain IDs
- Added references/dexscreener-api.md: pair endpoint + risk signal table
- No API key required for any data source
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant