Back to skill

Security audit

BlindOracle — Counterparty-Risk Checks for Agent-to-Agent Trades

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: the documented external API use appears purpose-aligned, though the scan was partial and one external-transmission finding remains noted.

Before installing, consider whether you are comfortable with the skill contacting api.craigmbrown.com and potentially sending the subject of a reputation lookup, and with paid calls. Confidence is medium: Skillspector completed execution but marked its analysis partial because several local path-like references could not be resolved, so the evidence is incomplete.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 51)May include surrounding context.

md
| `GET /v1/services` | every SKU with its live price and free-tier flag |
| `GET /v1/risk/payee?address=0x…` | OFAC SDN screen of an EVM address against a table rebuilt daily; a `no_match` is **not** a clearance |
| `GET /v1/delegations/{id}` | public view of a delegation chain |
| MCP `tools/list` on `https://api.craigmbrown.com/v1/mcp` | the same catalog as MCP tools, with `readOnlyHint` set honestly |

`agent.trust-badge` and `reputation.lookup` allow 5 free calls per day per caller
over MCP; a call that carries a payment is charged.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 74)May include surrounding context.

How a paid call works

text
POST https://api.craigmbrown.com/v1/services/reputation.lookup
Content-Type: application/json
{"subject": "<agent name or erc8004 id>"}
→ 402 with the exact price and the x402 payment requirements