Skip to content

BUG_BOUNTY.md

Latest commit

 

History

History
180 lines (131 loc) · 6.73 KB

BUG_BOUNTY.md

File metadata and controls

180 lines (131 loc) · 6.73 KB
Raw

Bug Bounty Program

UltrafastSecp256k1 -- Vulnerability Disclosure & Recognition

⚠️ Funding status: this program is NOT currently funded. There is no monetary bounty budget at this time. Valid reports receive public credit and a CVE (see §2). The reward figures in §2 are illustrative and aspirational — they would apply only if and when sponsorship is secured, entirely at maintainer discretion, and are not a commitment to pay. This matches SECURITY.md ("Funded bug bounty program — seeking sponsors"). Responsible reports are genuinely appreciated regardless; we will credit you in the advisory and CVE.

1. Scope

1.1 In-Scope Components

Component Priority Description
Core arithmetic (field/scalar/point) Critical Incorrect computation -> key recovery, forgery
ECDSA signing/verification Critical RFC 6979 nonce, signature correctness
Schnorr signing/verification Critical BIP-340 compliance
Constant-time layer (ct::) Critical Timing side-channel leaks
MuSig2 protocol High Key aggregation, rogue-key, nonce handling
FROST protocol High DKG, threshold signing, malicious participants
BIP-32 HD derivation High Child key derivation correctness
Address generation (27-coin) High Wrong addresses -> fund loss
C ABI (ufsecp) Medium NULL crashes, buffer overflows, UB
SHA-256 / RIPEMD-160 Medium Hash correctness
Serialization (DER, pubkey) Medium Parse confusion, malformed input crashes

1.2 Out-of-Scope

  • GPU backends (CUDA/OpenCL/Metal/ROCm) -- separate program planned
  • Language bindings (Python/Rust/Go/C#) -- report upstream to binding repos
  • Documentation errors (report as regular issues)
  • Denial-of-service via large input (not a security library concern at this layer)
  • Issues in dependencies (report to upstream maintainers)
  • Social engineering, phishing
  • Bugs in test code only (no production impact)

2. Severity Tiers & Aspirational Reward Ranges

The dollar ranges below are illustrative and contingent on the program being funded (see the funding-status notice at the top). They are not a current payment commitment. Until the program is funded, valid reports receive public credit + CVE.

Severity Description Reward Range (if/when funded)
Critical Private key recovery, signature forgery, nonce leak, CT bypass enabling key extraction $2,000 - $10,000
High Incorrect arithmetic producing wrong but non-exploitable results, FROST/MuSig2 protocol break, BIP-32 derivation error $500 - $2,000
Medium Crash from crafted input, memory safety issue (OOB read/write), UB in production code path $100 - $500
Low Non-security correctness bug (e.g., address checksum), minor API contract violation $50 - $100
Informational Documentation gaps, hardening suggestions, code quality improvements Public acknowledgment

2.1 Bonus Multipliers

Condition Multiplier
Includes working exploit / proof-of-concept 2x
Includes regression test case 1.5x
Affects CT layer with demonstrated timing measurement 2x
Found via formal methods / proof 1.5x

2.2 Recognition & Reward Conditions

  • First reporter of a unique vulnerability receives public credit (and a reward only if/when the program is funded — see the funding-status notice above)
  • Duplicate reports: first valid submission wins
  • Partial findings that lead to a full vulnerability are recognised proportionally
  • Any reward is at maintainer discretion and contingent on available funding
  • If funded, payment via cryptocurrency (BTC, ETH) or other agreed method

3. Reporting Process

3.1 How to Report

Do NOT open a public issue.

  1. GitHub Security Advisories (preferred):
    Create advisory

  2. Email: payysoon@gmail.com
    Encrypt with PGP if available (key published in SECURITY.md)

3.2 Report Requirements

A valid report must include:

  • Summary: One-paragraph description of the vulnerability
  • Affected component: File path(s) and function name(s)
  • Severity assessment: Your classification with justification
  • Reproduction steps: Minimal code or commands to trigger
  • Impact analysis: What an attacker could achieve
  • Environment: OS, compiler, version, build flags used

3.3 Response Timeline

Event SLA
Acknowledgment <= 72 hours
Severity triage <= 7 days
Fix timeline communicated <= 14 days
Critical fix released <= 30 days
High fix released <= 60 days
Medium/Low fix released Next minor release
Public disclosure 90 days after fix, or coordinated with reporter

4. Disclosure Policy

4.1 Coordinated Disclosure

  • Reporter agrees to keep findings confidential until public disclosure date
  • Maintainer agrees to:
    • Credit the reporter (unless anonymity requested)
    • Publish fix before or simultaneously with disclosure
    • Include reporter in advisory drafting (optional)

4.2 Safe Harbor

  • Good-faith security research is welcome
  • We will not pursue legal action against researchers who:
    • Follow this disclosure process
    • Do not access, modify, or delete user data
    • Do not disrupt services
    • Provide sufficient time for fix before disclosure

4.3 CVE Assignment

  • Critical and High findings will receive CVE identifiers
  • CVE will be requested through GitHub Security Advisories
  • Advisory will be published on GitHub and linked from SECURITY.md

5. Hall of Fame

Security researchers who responsibly disclose vulnerabilities will be credited in:

  • SECURITY.md Hall of Fame section
  • GitHub Security Advisory acknowledgments
  • Release notes for the fix version

To opt out of public credit, state so in your report.

6. Exclusions

The following are NOT eligible for rewards:

  • Issues already known and documented (check existing advisories)
  • Issues in pre-release/experimental code explicitly marked as such
  • Self-XSS, clickjacking, or UI-level issues (not applicable)
  • Rate limiting or brute force (not applicable at library level)
  • Findings that require physical access to the machine
  • Compiler bugs (report to compiler vendor)
  • Issues that only manifest with undefined build flags or non-supported configurations

7. Program Updates

This program may be updated at any time. Changes will be:

  • Committed to this file in the repository
  • Announced in release notes if reward tiers change
  • Applied prospectively (existing reports under previous terms)

Program effective date: 2026-02-24
Last updated: 2026-02-24