mirku
@mirku21 • 1,356 subscribers
21 | tech-ai | ongoing research
Videos
holy sh*t. Official Anthropic tip for Claude Code: stop burning Opus 5.5 context on tasks Haiku 5.5 can swarm, while Sonnet 5.5 builds put the architect on call with /advisor claude --advisor opus --subagents haiku Sonnet 5.5 drives the primary session on high effort, writing diffs and running test suites Haiku 5.5 subagents swarm the repo in parallel at 340 tok/s, handling file discovery and spec docs lookup Opus 5.5 stays on call in the background as the advisor, only stepping in at three critical moments: → before a plan locks: does the implementation plan miss auth invariants or schema contracts? → when a test breaks twice: are we fixing the root cause or falling into a recursive rabbit hole? → before calling done: did the full diff introduce hidden regressions or break pre-flight rules? Opus 5.5 reviews. Sonnet 5.5 builds. Haiku 5.5 swarms Jev engineering runs the identical pattern one layer down: mechanical decisions that need no reasoning (which file to open, which tool to invoke, retry or abort) execute in 16ms, so the frontier models only wake up when execution paths actually diverge Plan on high. Delegate on medium. Keep Opus on call. - the full advisor setup > Opus 5.5 on call reads full session history and catches deep architectural traps > Sonnet 5.5 lead drives edits, writes core logic, and executes test harnesses > Haiku 5.5 subagents swarm AST parsing, grep scans, and API docs in parallel > JEV micro-fork layer resolves 1,500+ mechanical routing branches in under 16ms > advisor stays completely silent on routine bash execution to protect context paste the setup and this prompt into Claude Code below: "Configure my Claude Code workspace for hierarchical advisor orchestration: 1. Audit ~/.claude/settings.json and project config for model roles fitting lead, subagents, and advisor: > Pin main session model: sonnet with effortLevel: high for primary execution > Pin subagents model: haiku with effortLevel: medium for parallel file discovery and docs retrieval > Pin advisor model: opus on call for strategic review 2. Enable the advisor tool in ~/.claude/settings.json: > Set advisorModel to claude-opus-5-5 > Enable subagent parallel dispatcher pool (3x workers) 3. Configure automatic advisor consultation checkpoints in ~/.claude/CLAUDE.md: > Consult /advisor opus before finalizing multi-file architecture plans > Automatically summon /advisor opus when the same test or compiler error fails twice > Enforce advisor diff contract audit before declaring tasks complete or staging git commits 4. Constrain subagent scopes: > Haiku subagents return structured AST summaries and docs snippets only without modifying main session context Show every configuration diff first. Do not apply edits until confirmed" ↳
mirku669,501 次观看 • 1 天前
this is pure f*cking treasure Claude Code tip: once Opus 5.5 is your main model, stop leaving Fable 5.1 sitting idle and stop burning Opus tokens on tasks Sonnet 5.5 can swarm put it on call with /goal claude --auto-mode "/goal all tests pass and lint is clean" Opus 5.5 keeps writing the code in auto mode Sonnet 5.5 swarms the routine work at medium Fable 5.1 inspects the full transcript and only speaks up at the turn boundary: → not yet met: am I drifting from acceptance criteria? → met: do the tests, diffs, and lint hold? → impossible: am I chasing an unreachable state? Fable grades. Opus ships JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or halt) go to JEV in under 16ms, and the big models only see the ones that split — the full tree > Opus 5.5 on high runs the main session > worker edits and patches code on Sonnet 5.5 > explorer indexes AST maps and call sites on Sonnet 5.5 > verifier runs tests and linters on Sonnet 5.5 > all three subagents on medium effort > Fable 5.1 on call as the Stop hook evaluator > JEV routes micro-forks under 16ms paste the tree and this prompt into Claude Code ↓ "Reconfigure my Claude Code setup around this tree: 1. Set the main session to Opus 5.5 on high effort with auto mode enabled. 2. Put Fable 5.1 on call as a session-scoped evaluator attached to the Stop hook: - Read-only transcript evaluation with zero tool calls - Three structured outcomes: not yet met (with steering notes), met (auto-clear), and impossible (abort early) 3. Spawn Sonnet 5.5 subagents on medium effort for worker, explorer, and verifier roles. Defer goal checks while background tasks run. 4. Route fast deterministic forks (tool resolution, path selection, auto-retries) through JEV. 5. Add one rule to CLAUDE.md: run long tasks with: claude --auto-mode \"/goal \" Show me every configuration change as a diff first. No edits until I say go." ↳
mirku40,611 次观看 • 2 天前
没有更多内容可加载