Agent skill
protocol-templates
Structured, protocol-type-specific audit templates enumerating the exact checks, invariants, and attack vectors relevant to each protocol category. Use when auditing AMM/DEX, lending, bridge, governance, or vault protocols to load targeted checklists based on context detection results.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/protocol-templates
SKILL.md
Protocol Templates Skill
Purpose
Provide structured, protocol-type-specific audit templates that enumerate the exact checks, invariants, and attack vectors relevant to each protocol category. These templates are loaded based on context detection.
Available Templates
| Template | Protocol Type | Key Focus Areas |
|---|---|---|
| AMM/DEX | Uniswap, Curve, Balancer-style | Price manipulation, LP attacks, MEV |
| Bridge | Cross-chain bridges | Message verification, replay, accounting |
| Lending | Aave, Compound-style | Oracle, liquidation, interest rates |
| NFT Marketplace | OpenSea, Blur-style | Order validation, royalties, signatures |
| Staking | Lido, RocketPool-style | Reward distribution, withdrawal, delegation |
Template Structure
Each template follows a consistent format:
- Protocol Overview: What the protocol does
- Architecture Checklist: Structural checks
- Invariants: Mathematical properties that must hold
- Attack Vectors: Known exploits for this protocol type
- Critical Functions: Functions requiring deepest review
- Integration Risks: External dependency risks
- Economic Considerations: Game theory and incentive analysis
Usage
- Context detection identifies protocol type
- Appropriate template is loaded
- Auditor follows template checklist systematically
- Findings tagged with template category for standardized reporting
Prerequisites
Protocol templates require Context Detection for automatic loading. Manual template selection is also supported.
Validation
To verify template coverage, validate against known protocol types:
# Validate template completeness
required_sections = ["Overview", "Architecture Checklist", "Invariants", "Attack Vectors", "Critical Functions"]
for template in templates:
for section in required_sections:
assert section in template.sections, f"{template.name} missing {section}"
# Example template loading configuration
detected_type: amm-dex
confidence: 0.92
loaded_templates:
- amm-dex-template.md
- defi-patterns.md
checklist: dex-amm-checklist
# Test template file integrity
for f in *-template.md; do echo "Validating $f"; head -5 "$f"; done
Behavior Guidelines
- Template loading is required when context detection confidence exceeds 0.8
- Auditors MUST verify that all invariants in the template are tested
- Custom template sections may optionally be added for novel protocol designs
References
- Protocol Templates References - Template customization guides and architecture patterns
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?