X · by Movez
A 12-page PDF on building a Jev harness
A 10-step blueprint for coding agents, credited to the Jev founder.
Movez
@0xMovez
Jev Founder, Diogo Amogo, just released 12-page PDF on building a Jev Harness for coding agents this is a 10-step blueprint on how to make your coding agents 200× faster and 400× cheaper: step 1 → meet Jev: the LLM writes, the harness executes, Jev decides what each turn sees, where it routes and whether it runs step 2 → ask the question that breaks every agent: how would you design one if LLMs had no KV cache? step 3 → stop routing blind: Opus → Sonnet → Opus costs 6.19 vs 4.15 for pure Opus, because handing back reprocesses the whole context step 4 → follow the tokens: reading and searching take 56.2% of tool turns and 46.5% of tokens. Writing code is under 10% step 5 → score every chunk per query: hide, short summary, long summary or full. Compress after the question, not before step 6 → disclose tools in tiers: one-line snippets for 100s of tools, schema on demand, docs for one-off queries. Batteries stop costing context step 7 → load instructions by condition: touching *.tsx loads the style guide, billing/ loads its gotchas file, and compaction can't erase either step 8 → route by trust, not just difficulty: secrets and infra stay on first-party frontier models, public docs go to the cheapest one step 9 → share one retrieval pass: cross-model review, eval generation, ELI5 explainers and live progress pages all run read-only in the background step 10 → gate every command: programmable allow / ask / deny policies that read a script before it runs, not just its name Send this PDF and the article below to your Claude Code or Codex instance and start shipping 200× faster.
Sep 21, 2026 · 2.0K likesOpen on X
Movez shares a 12-page PDF, credited in the post to Jev's founder, on building a Jev harness for coding agents. The LLM writes, the harness executes, and Jev decides what each turn sees, where it routes and whether it runs.
The post sums the PDF up as ten steps: design as if LLMs had no KV cache, stop routing blind between models, score every context chunk per query, disclose tools in tiers, load instructions by condition, route by trust as well as difficulty, share one retrieval pass, and gate every command with allow, ask and deny policies.
It claims coding agents get up to 200× faster and 400× cheaper, and quotes Opus → Sonnet → Opus costing 6.19 against 4.15 for pure Opus, because handing back reprocesses the whole context. Reading and searching take 56.2% of tool turns and 46.5% of tokens; writing code is under 10%. These are the post's figures.
Open the source
