The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various CLI tools for diagnostics and environment management.
Evidence: Use of vercel ls, vercel logs, vercel env ls, and vercel env add for platform interaction.
Evidence: Use of git log and git diff for version control analysis.
Evidence: Use of npx tsc --noEmit for TypeScript validation.
Evidence: A shell loop in the 'Quick fix' section for Section B reads .env.local and executes vercel env add for each entry to sync local secrets to the production environment.
[DATA_EXFILTRATION]: The skill facilitates the transfer of sensitive local configuration data to a remote platform.
Evidence: The skill reads the contents of .env.local and sends the keys and values to the Vercel production environment via the vercel env add command. This is documented as a standard diagnostic and sync procedure within the Vercel ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests and processes untrusted data from external sources.
Ingestion points: The agent reads data from vercel logs (runtime/build output) and vercel env ls (environment variable keys).
Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions that might be embedded in log outputs or environment variable values.
Capability inventory: The skill has access to shell command execution (vercel, git, npx), file reads (cat, grep), and network operations via curl and platform CLI tools.
Sanitization: No explicit sanitization or validation of log content or environment variables is mentioned before processing.

vercel-supabase-debugger