Automate batch LLM calls and Lean4 theorem proving with JSON workflows.
npx skills add https://github.com/grahama1970/fetcher --skill scillm
Please help me install this Agent Skill. Skill: scillm Source: https://github.com/grahama1970/fetcher/tree/main/.agents/skills/scillm Command: npx skills add https://github.com/grahama1970/fetcher --skill scillm
Automates batch LLM calls and Lean4 theorem proving to accelerate complex reasoning tasks.
Invoke batch.py with prompts to run multiple completions in parallel and use prove.py to perform Lean4 proofs.
High-intent search queries and answers about installing and using this skill.
Batch LLM calls and Lean4 proofs run together by executing parallel completions and formal verifications through a CLI and Python API, allowing multiple prompts and proofs to process simultaneously for complex reasoning tasks.
Yes, you can run Lean4 proofs in parallel with LLM completions. Batch mode executes multiple prompts and formal proofs simultaneously, enabling automated reasoning and verification within a single workflow.
Yes, batch LLM and Lean4 proof automation requires CHUTES_API_BASE, CHUTES_API_KEY, and CHUTES_MODEL_ID environment variables to access the paved-path APIs needed for execution.
The best way to integrate batch LLM completions into an automation pipeline is using the importable Python API, which supports single and batch operations with optional JSON responses for downstream consumption.
Yes, batch LLM processing supports optional JSON output. Both single and batch modes offer JSON formatting, allowing completions to be directly consumed by subsequent automation steps.
Lean4 formal verification ensures generated proofs are logically sound and verifiable, providing mathematical certainty that LLM completions alone cannot guarantee for complex reasoning outputs.
Abdullahmohammadareftest1
parcadeiIndex artifacts immediately upon creation using event-driven triggers.
co-labs-coAutomate design and testing of Python CLI apps with Click and Rich.
zapabobTrain SFT+GRPO models for mathematical theorem proving with MCP/A2A integration.
