ClawAudit verdict
prototype-design
Receives external input AND uses eval
Skill generates HTML UI prototypes locally using design templates from a references directory; all work is local file generation with no external network calls or suspicious behavior.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
What it does
These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Receives external input AND uses eval — the remote code-injection pattern (data-flow not verified)
LLM01 · LLM05 · ASI01 · ASI05
Reads files AND uses eval/Function — may execute locally stored payloads
LLM05 · LLM06 · ASI05
Permission integrity
file_read
Findings (4)
<script> tag in markdown — potential code injection
SKILL.md · code · <script>
Instructs covert action — may act without user awareness
references/design-systems/claude-DESIGN.md · prose · downgraded · quietly
Dynamic import() — loads module at runtime
references/design-systems/linear.app-DESIGN.md · prose · downgraded · import("
Python directory traversal
SKILL.md · code · os.listdir(
Permissions & capabilities
No declared permissions — minimal attack surface.
file_readdir_traversalnetwork_indynamic_eval Thanks — recorded.