1inch
Verdict safe, but 2 findings (2 medium), mainly data exfiltration, social engineering. Most recent scan 2026-02-05.
This skill provides legitimate 1inch DEX aggregator functionality but has MEDIUM severity concerns around financial transparency and embedded fee collection. The hardcoded referral wallet address (0x890CACd9dEC1E1409C6598Da18DC3d634e600b45) receives 0.3% of all swap values, creating a financial incentive structure that may not be sufficiently disclosed to end users. While the fee is documented, it is not prominently featured in the skill description and users cannot easily opt out. The skill does not contain malicious code, data exfiltration, or prompt injection attempts. The primary concerns are around social engineering through insufficient disclosure of the fee structure and the hardcoded nature of the beneficiary address. Additional LOW severity issues include missing optional metadata fields and lack of error handling guidance. Recommended actions: enhance fee disclosure, make referral configuration optional, and add user consent mechanisms.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
1inch DEX aggregator. Find the best swap rates across 400+ liquidity sources on 12+ blockchains.
- Package
- 2 files, 6.4 KB
- SKILL.md
- 677 words, 12+ sections
- Code
- None. Instructions only.
- Requires
- bins:
curljq
SKILL.md covers: 1inch DEX Aggregator 🦄 · Environment Variables · 💎 Referral Fee Configuration · Features · API Base URL · Get Swap Quote · Token addresses · Referral configuration · Get Quote Only (No Transaction) · Fusion Mode (Gasless Swap) · Get Fusion quote · Get Token List
Findings
MEDIUM Hardcoded Referral Wallet Address for Fee Collection Data Exfiltrationllm
The skill contains a hardcoded Ethereum wallet address (0x890CACd9dEC1E1409C6598Da18DC3d634e600b45) configured to receive 0.3% referral fees from all swaps. While this is disclosed in the documentation, users may not realize that every transaction executed through this skill automatically sends fees to this third-party address. This creates a financial incentive structure that is embedded in the skill's configuration and cannot be easily modified by users.
SKILL.md
REFERRER_ADDRESS: 0x890CACd9dEC1E1409C6598Da18DC3d634e600b45, FEE_PERCENT: 0.3MEDIUM Insufficient Disclosure of Financial Incentive Structure Social Engineeringllm
While the skill documentation mentions the 0.3% referral fee, the disclosure is buried in a table format and may not be sufficiently prominent for users to understand that they are paying fees on every transaction. The skill description ('Find the best swap rates') does not mention that rates include a mandatory 0.3% fee to a third party, which could be considered misleading by omission.
SKILL.md
Description: '1inch DEX aggregator. Find the best swap rates across 400+ liquidity sources on 12+ blockchains.' - No mention of embedded fees in primary descriptionWhat is in the package
| File | Kind | Size |
|---|---|---|
README.md | doc | 714 B |
SKILL.md | doc | 5.7 KB |
External hosts referenced
api.1inch.dev 1inch.io clawdbot.com docs.1inch.io portal.1inch.dev
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Environment variables referenced
CHAIN_ID API_KEY DST_TOKEN SRC_TOKEN ONEINCH_API_KEY TOKEN_ADDRESS
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/1inch
- Raw scan result: 1inch-scan.json
- Badge for your README:
- Request a re-scan