Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed, read-only DeFi research skill; it sends public wallet and research data to FarmDash when used, but does not sign, trade, or move funds.
Install only if you are comfortable sending DeFi research queries, public wallet addresses, token/chain parameters, and any optional FarmDash API key to FarmDash. Do not provide private keys, seed phrases, signed messages, or wallet exports. Treat route links and partner disclosures as commercial recommendations, verify destination domains independently, and use a separate execution tool only after explicit confirmation.
The usage cue includes the phrase "where should I deploy capital next?", which is broad natural language that could plausibly appear in general financial discussions outside the intended FarmDash context. The file does not pair this example with explicit negative examples or tighter trigger constraints, so it may increase unintended invocation risk in agents that rely on description text.
The document asserts that its 'Data Sent to FarmDash MCP' table is the complete list of transmitted data, but later introduces additional transmissions: an onboarding POST with agentAddress and a universal X-ClawHub-Skill header. This creates an incomplete disclosure pattern that can cause operators to underestimate external data exposure and violates the principle of transparent data minimization.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
* **Time-bound analysis.** "Based on data as of [timestamp]." DeFi changes fast.
* **Match constraints, not labels.** "Aggressive" never waives liquidity, data-quality, protocol, stablecoin, bridge, or concentration limits.
* **Keep the handoff transparent.** The agent's job is to present complete, honest information - including limits, fees, downside cases, destination verification, and the `https://www.farmdash.one/fees` disclosure when a FarmDash route is shown. The user still decides whether to sign or fund any action through a separate execution surface.
* **Refuse harmful requests.** If asked to bypass user confirmation, store secrets, or push a recommendation that the data does not support, refuse and explain.
## Data Sources
* **DeFiLlama:** TVL, protocol metrics
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
* FarmDash may receive referral, affiliate, or routing compensation when users follow supported FarmDash routes - always disclosed with `https://www.farmdash.one/fees`.
* The user retains full responsibility for every on-chain decision.
**Install:** Copy this file into your OpenClaw workspace, or fetch `https://www.farmdash.one/openclaw-skills/farmdash-trail-intelligence/SKILL.md`.
### Companion skills:
* FarmDash Wagon Steward — read-only portfolio aggregation + capital efficiency
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
* FarmDash may receive referral, affiliate, or routing compensation when users follow supported FarmDash routes - always disclosed with `https://www.farmdash.one/fees`.
* The user retains full responsibility for every on-chain decision.
**Install:** Copy this file into your OpenClaw workspace, or fetch `https://www.farmdash.one/openclaw-skills/farmdash-trail-intelligence/SKILL.md`.
### Companion skills:
* FarmDash Wagon Steward — read-only portfolio aggregation + capital efficiency
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
* FarmDash may receive referral, affiliate, or routing compensation when users follow supported FarmDash routes - always disclosed with `https://www.farmdash.one/fees`.
* The user retains full responsibility for every on-chain decision.
**Install:** Copy this file into your OpenClaw workspace, or fetch `https://www.farmdash.one/openclaw-skills/farmdash-trail-intelligence/SKILL.md`.
### Companion skills:
* FarmDash Wagon Steward — read-only portfolio aggregation + capital efficiency
The skill metadata and repeated prose claim the capability 'never executes' and is strictly read-only, but the document later instructs the operator to perform an outbound POST onboarding call and to attach a tracking header on requests. Even if no blockchain transaction occurs, this is still active network side-effect behavior and contradicts the stated trust boundary, which can mislead users about what data is sent and when.
The onboarding example explicitly sends an agent wallet address to an external service via POST. Although framed as optional and consent-based, it is still external data transmission that can create privacy, correlation, and tracking risk, especially because wallet addresses are durable identifiers in a sensitive financial context.
Optional command after consent:
curl -X POST https://www.farmdash.one/api/v1/agent/onboard \
-H "Content-Type: application/json" \
-H "X-ClawHub-Skill: farmdash-trail-intelligence" \
-d '{"agentAddress": "0xYOUR_AGENT_WALLET"}'
The instruction "Use this skill when choosing where to farm" describes a general situation rather than a precise invocation boundary. While the surrounding document is detailed, this specific activation wording does not clearly distinguish when the skill should not be used versus other research or execution skills.
No suspicious patterns detected.