sandeco-token-reduce-ollama

sandeco-token-reduce-ollama is a Prompt design & optimisation skill for AI agents, published by sandeco in the prompts repository on GitHub. Skillspace rates sandeco-token-reduce-ollama F (45/100) and the safety scan rated it risky; the rating is based on static analysis only.

Risky. This skill has patterns associated with unsafe behaviour. Read the source carefully before running it.

Comprime tokens de prompts usando reescrita generativa com um modelo Ollama local (default: qwen3:8b) para reduzir custo e latencia ao enviar ao Claude. Aceita texto direto, arquivo .txt ou PDF como entrada: extrai o PDF para markdown (via pymupdf4llm), salva o .md ao lado do PDF e comprime o markdown. Use esta skill sempre que o usuario pedir para comprimir/reduzir/economizar tokens E mencionar Ollama, qwen, qwen3, llm local, modelo local, rodar localmente, "sem BERT", ou quiser compressao via LLM generativo em vez de classificador. Tambem use quando o usuario quiser pre-processar contexto localmente (incluindo PDFs) antes de enviar ao Claude usando Ollama. Triggers: "comprimir com ollama", "comprimir localmente", "ollama compress", "qwen3", "qwen", "reescrever com ollama", "resumir com llm local", "compressao generativa", "pdf com ollama", "ler pdf com ollama", "inicializar token-reduce-ollama", "init token-reduce-ollama".

Grade

F

Quality

45 / 100

Safety

Risky

GitHub stars

863

Unproven: rated by static analysis only, not yet compared head-to-head.

Our two independent classifiers differ on this skill's category, so treat the placement as provisional.

This rating describes the version of the repository we crawled; it may have changed since.

Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.