Access DeepSeek chat, reasoning, and code generation via curl-based API requests.
npx skills add https://github.com/vm0-ai/vm0-skills --skill deepseek
Please help me install this Agent Skill. Skill: deepseek Source: https://github.com/vm0-ai/vm0-skills/tree/main/deepseek Command: npx skills add https://github.com/vm0-ai/vm0-skills --skill deepseek
This Skill provides curl-based access to OpenAI-compatible DeepSeek models for chat, reasoning, and code generation, enabling cost-effective AI integration.
Basic chat: bash -c 'curl -s "https://api.deepseek.com/chat/completions" -X POST -H "Content-Type: application/json" -H "Authorization: Bearer ${DEEPSEEK_API_KEY}" -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"Hello"}]}' | jq .'
High-intent search queries and answers about installing and using this skill.
DeepSeek provides OpenAI-compatible API endpoints. Use curl with your API key in the Authorization header and POST to https://api.deepseek.com/chat/completions with a JSON payload specifying the model (deepseek-chat or deepseek-reasoner) and messages. The API returns completions in OpenAI format, supporting streaming and standard response modes.
DeepSeek-chat handles general conversation and code generation; deepseek-reasoner provides deep reasoning capabilities; and fill-in-the-middle (FIM) workflows support code completion. Each model is optimized for its task and accessible through the same OpenAI-compatible endpoint with curl.
Yes. DeepSeek supports code generation through deepseek-chat and fill-in-the-middle (FIM) mode. Configure temperature, top_p, and max_tokens via curl payload parameters to control generation behavior. Multi-turn conversations enable iterative code refinement and function-calling integration.
Control output with temperature (randomness), top_p (nucleus sampling), max_tokens (response length), and streaming mode. Set these in the JSON request payload sent via curl. Error handling manages failed requests; function-calling support enables tool integration for extended capabilities.
Yes. DeepSeek exposes OpenAI-compatible endpoints, so curl requests follow standard OpenAI API structure. This compatibility enables drop-in replacement for tools and workflows expecting OpenAI format, using the same authentication and payload patterns.
openshift-engRetrieve OAuth tokens and inject Authorization headers into curl requests for OpenShift clusters.
disuseredParse, filter, and transform JSON data with jq queries.
kitaekattAuthor, validate, and execute declarative YAML workflows with shell and model nodes.
dennisonbertramOrchestrate local LLM inference and model management with Ollama.
