Back to skill

Security audit

Coil Board

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed market-data helper that reads Coil's public or paid API endpoints and does not show hidden execution, persistence, credential theft, or trading actions.

Install only if you are comfortable with an agent reading market research from coil.trade and, for live paid slices, using an x402-capable wallet/client to make small USDC payments. Treat the output as research, not trading instructions, and confirm whether your agent is using the free delayed script, a free key, the remote MCP server, or paid live endpoints before relying on freshness.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill advertises paid x402 access to a live, scored board but the analyzed behavior reportedly only uses free public endpoints and omits the claimed payment, full-board, and score fields. That mismatch can mislead downstream agents into making financial decisions on delayed or incomplete data while believing they received the authoritative paid dataset, creating integrity and decision-quality risk.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

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.

Content

Scanner excerpt · .github/workflows/ci.yml (reported line 18)May include surrounding context.

yaml
# /.well-known/agent-skills/index.json advertises). A repo copy that drifts
        # is how a skill starts telling people things that are no longer true.
        run: |
          curl -sS https://coil.trade/skills/coil-board/SKILL.md -o /tmp/live.md
          diff -u /tmp/live.md SKILL.md || {
            echo "::error::SKILL.md has drifted from https://coil.trade/skills/coil-board/SKILL.md"
            exit 1; }

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

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.

Content

Scanner excerpt · .github/workflows/ci.yml (reported line 20)May include surrounding context.

yaml
# /.well-known/agent-skills/index.json advertises). A repo copy that drifts
        # is how a skill starts telling people things that are no longer true.
        run: |
          curl -sS https://coil.trade/skills/coil-board/SKILL.md -o /tmp/live.md
          diff -u /tmp/live.md SKILL.md || {
            echo "::error::SKILL.md has drifted from https://coil.trade/skills/coil-board/SKILL.md"
            exit 1; }

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

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.

Content

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

md
## Canonical source

`SKILL.md` here is a byte-identical mirror of
<https://coil.trade/skills/coil-board/SKILL.md>, which is what
`/.well-known/agent-skills/index.json` publishes a sha256 for. **CI fails if this copy drifts** —
a skill that quietly diverges from its source is how one starts telling people things that are no
longer true.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.