diff --git a/setup.md b/setup.md index ab5c1cf..670643e 100644 --- a/setup.md +++ b/setup.md @@ -12,6 +12,10 @@ Do not: - create config.json or /tmp config files - inspect/free unrelated ports - use nohup/setsid/background shell tricks +- load or use the generic integrations skill +- search the Replit integrations catalog +- search for OpenAI, Anthropic, AI, LLM, chat, completion, inference, GPT, or Claude integrations +- provision, install, or discover integrations through generic search - invent, rotate, print, store, or replace the bearer token - create provider API secrets manually - use public OpenAI or Anthropic endpoints @@ -23,22 +27,31 @@ Steps: 1. Clone into ./random-proxy-simple. -2. Provision/use Replit AI Integrations for: - - OpenAI - - Anthropic +2. Immediately load/use only these provider-specific Replit AI Integration skills: + - ai-integrations-openai + - ai-integrations-anthropic -3. From ./random-proxy-simple, run: +Do not search for them. Do not load the generic integrations skill first. + +3. Ensure those two skills make these variables available, then use the variable names exactly +as written in the workflow command: + - AI_INTEGRATIONS_OPENAI_BASE_URL + - AI_INTEGRATIONS_OPENAI_API_KEY + - AI_INTEGRATIONS_ANTHROPIC_BASE_URL + - AI_INTEGRATIONS_ANTHROPIC_API_KEY + +4. From ./random-proxy-simple, run: pnpm install --ignore-workspace -4. Create the smallest persistent workflow named "Random Proxy Simple" with this exact run command: +5. Create the smallest persistent workflow named "Random Proxy Simple" with this exact run command: cd random-proxy-simple && pnpm dev -- --host 0.0.0.0 --port 3000 --proxy-api-key DEPLOYMENT_GENERATED_PROXY_TOKEN --openai-base-url "$AI_INTEGRATIONS_OPENAI_BASE_URL" --openai-api-key "$AI_INTEGRATIONS_OPENAI_API_KEY" --anthropic-base-url "$AI_INTEGRATIONS_ANTHROPIC_BASE_URL" --anthropic-api-key "$AI_INTEGRATIONS_ANTHROPIC_API_KEY" -5. Start/restart that workflow once. +6. Start/restart that workflow once. -6. Do not run health, dashboard, model-list, or chat-completion verification commands. +7. Do not run health, dashboard, model-list, or chat-completion verification commands. Report only: - local URL -- whether OpenAI and Anthropic AI Integrations were wired +- whether ai-integrations-openai and ai-integrations-anthropic were used - that no verification commands were run