技能详情(站内镜像,无评论)
作者:Rowan @0xRowan
许可证:MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本:v1.0.0
统计:⭐ 0 · 103 · 0 current installs · 0 all-time installs
⭐ 0
安装量(当前) 0
🛡 VirusTotal :良性 · OpenClaw :良性
Package:0xrowan/onchain-contract-token-analysi
安全扫描(ClawHub)
- VirusTotal :良性
- OpenClaw :良性
OpenClaw 评估
The skill is an instruction-only onchain contract/token analysis guide whose required resources and instructions are consistent with its stated purpose.
目的
Name and description match the SKILL.md instructions. The skill requests no binaries, env vars, or installs and is coherent for doing contract/token security analysis.
说明范围
Instructions ask the agent to infer from 'files, addresses, ABI names, deployment scripts, or docs' and to verify with live chain/explorer data when needed. That scope is appropriate for an analysis skill but broad — it implies reading any provided source artifacts and making external network queries to block explorers or nodes.
安装机制
No install spec and no code files. Lowest-risk delivery model (instruction-only).
证书
The skill requires no environment variables or credentials up front, which is proportional. Practically, effective use may require user-provided artifacts (ABI/source), contract addresses, or read-only node/explorer API keys (e.g., Etherscan/Alchemy) supplied at runtime — the SKILL.md does not request secrets, but users may choose to provide tokens outside the skill manifest.
持久
always:false and the skill is user-invocable. It does not request persistent privileges or modify other skills or system settings.
综合结论
This skill is a coherent, instruction-only analyzer — it doesn't install software or ask for credentials by itself. Before using it: only provide the contract addresses, ABIs, or source files you intend to share; do not paste private keys or wallet JSONs. If you supply node or explorer API keys, prefer read-only keys and revoke them after use. Be aware the agent may perform or request live-chain lookups (block explorer or node queries) to conf…
安装(复制给龙虾 AI)
将下方整段复制到龙虾中文库对话中,由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库(龙虾 AI)执行:为本机安装 OpenClaw 技能「onchain contract token analysis」。简介:Analyze smart contracts, token mechanics, permissions, fee flows, upgradeabilit…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装:https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/0xrowan/onchain-contract-token-analysi/SKILL.md
(来源:yingzhi8.cn 技能库)
SKILL.md
---
name: onchain-contract-token-analysis
description: Analyze smart contracts, token mechanics, permissions, fee flows, upgradeability, market risks, and likely attack surfaces for onchain projects. Use when reviewing ERC-20s, launchpads, vaults, staking systems, LP fee routing, ownership controls, proxy setups, or suspicious token behavior.
user-invocable: true
metadata: {"openclaw":{"emoji":"🔍","skillKey":"onchain-contract-token-analysis"}}
---
# Onchain Contract / Token Analysis
Use this skill when the task is to assess a token, protocol, launch module, vault, staking system, router, or related onchain project from a security, permissions, tokenomics, or behavior perspective.
## Core objective
Produce a practical analysis that answers:
- What the system does
- Who controls it
- How value and fees move
- What privileged actions exist
- What users can lose money from
- Whether there are obvious red flags or design risks
## Workflow
### 1. Identify the scope
First determine which of these the request actually targets:
- token contract
- factory / launcher
- vault / staking / locker
- router / hook / proxy / module
- admin / governance / registry
- full protocol system
If the scope is unclear, infer it from the files, addresses, ABI names, deployment scripts, or docs.
### 2. Map the architecture
Before judging risk, build a compact model of the system:
- main contracts
- ownership / admin roles
- external dependencies
- upgradeability pattern
- event flow
- token creation flow
- fee routing flow
Prefer a short system map over long prose.
### 3. Check control and permissions
Always verify:
- `owner`, `admin`, `governor`, `operator`, `manager`, `signer`
- role-based access control
- pausable / blacklist / whitelist powers
- mint / burn / seize / rescue / withdraw permissions
- parameter setters
- upgrade authority
- emergency functions
Call out who can do what, and whether those powers are bounded or dangerous.
### 4. Check token mechanics
For ERC-20 and tokenized systems, verify:
- total supply model
- mintability
- burnability
- transfer restrictions
- fee on transfer / tax
- max wallet / max tx rules
- trading enable switch
- blacklist / antibot logic
- rebasing / reflection / hidden balance logic
- allowance edge cases
If the token claims to be standard, confirm whether behavior actually matches that claim.
### 5. Check fee and value flow
Trace where user funds or protocol fees go:
- LP fee recipients
- treasury recipients
- locker / vault recipients
- protocol fee splits
- conversion / swap path
- withdrawal path
- claim path
Do not just name recipients. Explain whether they are:
- immutable
- admin-changeable
- delayed
- claim-based
- dependent on offchain identity or signatures
### 6. Check upgradeability and mutability
If proxies or modules exist, verify:
- proxy type
- implementation admin
- initialization safety
- reinitialization protection
- storage layout assumptions
- upgrade trust model
If not upgradeable, still check whether behavior can change through configurable modules.
### 7. Check attack surface
Look for:
- arbitrary external calls
- reentrancy opportunities
- unchecked token callbacks
- unsafe approvals
- signature replay
- missing nonce / deadline checks
- address(0) edge cases
- misconfigured recipient logic
- accounting mismatch
- stale state after recipient updates
- rounding leakage
- griefing / denial-of-service vectors
When risk depends on business assumptions, state that explicitly.
### 8. Check market-facing risk
When the target is a token or launch flow, explicitly assess:
- honeypot-like behavior
- sell restrictions
- hidden tax changes
- admin ability to freeze exits
- liquidity custody
- locker guarantees
- whether front-end labels could misclassify the asset
Do not overclaim. Distinguish:
- confirmed malicious logic
- dangerous centralization
- poor design
- heuristic / market-behavior false positives
## Output format
Default to this structure:
### Summary
One short paragraph stating what the system is and the top conclusion.
### Findings
List issues in severity order:
- severity
- title
- affected contract / function
- why it matters
- exploit or failure mode
- whether it is confirmed or conditional
### Trust model
State:
- who controls upgrades
- who controls fees
- who controls pauses or restrictions
- what users must trust offchain
### Token / fee flow
Explain:
- how tokens are created
- where fees accrue
- who can claim them
- what can change later
### Open questions
List anything blocked by missing source, missing ABI, missing deployment info, or offchain dependencies.
## Special guidance
### When reviewing a suspicious token
Be precise:
- "can blacklist holders" is stronger than "looks risky"
- "owner can change tax" is stronger than "may be a scam"
- "no onchain sell restriction found" is stronger than "not a honeypot"
### When reviewing a launch module
Always distinguish:
- launcher logic
- underlying token implementation
- LP locker behavior
- fee locker behavior
- who receives economic rights
### When chain data is required
If the task depends on live state, verify with current chain or explorer data instead of assuming from source alone.
## Do not
- Do not call something malicious without code-based support
- Do not confuse admin centralization with exploitability
- Do not ignore offchain identity dependencies when they control payouts
- Do not stop at contract syntax; trace actual economic outcomes