Prompt Shield (Jailbreak / Injection Detection)

Classify a prompt before it reaches your LLM. Brainiall Prompt Shield engine. Returns category (jailbreak | prompt_injection | data_exfiltration | impersonatio

liveverifiedunknown~5s typical

mcpprobedbrainiallnlp

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-com-brainiall-nlp /detect_prompt_injection for me.

Input

Field	Type	Description
prompt*	string	The prompt text to classify (NOT executed)

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-com-brainiall-nlp","endpoint":"/detect_prompt_injection","input":{"prompt":"..."}}'

cli

npx cracked-ai run -p mcp-com-brainiall-nlp -e /detect_prompt_injection -i '{"prompt":"..."}'

mcp

run_tool({ provider: "mcp-com-brainiall-nlp", endpoint: "/detect_prompt_injection", input: {"prompt":"..."} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.