Trust Assessment
pumpclaw-base received a trust score of 68/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 4 findings: 0 critical, 1 high, 3 medium, and 0 low severity. Key findings include Missing required field: name, Unpinned npm dependency version, Skill requires and uses BASE_PRIVATE_KEY for blockchain transactions.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Skill requires and uses BASE_PRIVATE_KEY for blockchain transactions The skill explicitly requires the `BASE_PRIVATE_KEY` environment variable to operate, using it to create a `viem` wallet client for signing blockchain transactions. This grants the skill full control over the associated wallet, posing a significant risk if the skill or its execution environment is compromised. While necessary for the skill's intended functionality, this direct handling of a private key is a high-risk operation. Implement secure credential management practices for AI agents. Avoid storing private keys directly in environment variables in production. Consider using hardware security modules (HSMs), secure enclaves, or dedicated key management services. Ensure the agent's execution environment is isolated and hardened. The agent orchestrator should explicitly confirm user consent before allowing skills to access sensitive credentials. | LLM | scripts/pumpclaw.ts:18 | |
| MEDIUM | Missing required field: name The 'name' field is required for claude_code skills but is missing from frontmatter. Add a 'name' field to the SKILL.md frontmatter. | Static | skills/clawd800/pumpclaw-base/SKILL.md:1 | |
| MEDIUM | Unpinned npm dependency version Dependency 'viem' is not pinned to an exact version ('^2.45.1'). Pin dependencies to exact versions to reduce drift and supply-chain risk. | Dependencies | skills/clawd800/pumpclaw-base/scripts/package.json | |
| MEDIUM | Untrusted markdown contains shell command examples The `SKILL.md` (untrusted input) contains shell command examples (`export`, `npx tsx`). An AI agent interpreting this markdown could be susceptible to prompt injection if it is instructed to execute commands found within untrusted content without proper validation or sandboxing, potentially leading to arbitrary command execution on the host system. AI agents should be designed to never execute shell commands directly from untrusted input. All commands should be explicitly confirmed by the user or passed through a secure execution environment with strict allow-listing and parameter sanitization. The agent orchestrator should parse and validate commands from untrusted sources rather than executing them verbatim. | LLM | SKILL.md:18 |
Scan History
Embed Code
[](https://skillshield.io/report/5fed185c9cd02b1c)
Powered by SkillShield