meme-launcher
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external domains including pump.fun, raydium.io, jup.ag, and solscan.io via the
WebFetchandmcp__perplexity-ask__searchtools. Untrusted content from these third-party platforms could potentially contain malicious instructions designed to influence the agent's behavior. - Ingestion points: WebFetch domains (pump.fun, raydium.io, jup.ag, solscan.io, docs.solana.com) and perplexity search.
- Boundary markers: None identified in the skill instructions to delimit external data from agent commands.
- Capability inventory: The skill possesses the
TodoWritecapability and instructions to execute shell commands vianpx. - Sanitization: No explicit sanitization or filtering of external web content is described.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a pre-launch validation script using the command
npx tsx .claude/skills/meme-launcher/scripts/pre-launch-check.ts. This pattern involves running executable code provided within the skill's own directory structure.
Audit Metadata