token-efficiencylisted

Token optimization for Claude Code CLI sessions: RTK (Rust Token Killer) command-output compression, efficient file and command strategies, role-based model routing for subagents, Source-of-Truth keyword framework, hybrid SQLite agent memory with checkpoints, and concise output. Use when installing or verifying RTK or its hook, when a dev command's output is large (test runs, builds, installs, long logs), when spawning coding subagents, when writing agent logs or querying agent memory, when seeding a project's SoT keyword index, or when folding agent memory at a checkpoint. Also use when the user mentions 'rtk', 'token efficiency', 'token savings', 'context bloat', 'model routing', 'SoT keywords', or 'checkpoint agent memory', or asks for shorter or less verbose output, or to stop narrating or dumping whole files.
M4NUSH7/Niche-Claude-Code · ★ 1 · AI & Automation · score 77

Install: claude install-skill M4NUSH7/Niche-Claude-Code

# Token Efficiency **Check the steady state first.** If `rtk init --show` reports the hook is active and the `templates/CLAUDE-md-snippet.md` block is in CLAUDE.md, pillars 1, 3 and 5 are already enforced with this skill out of context - apply pillars 2 and 4 only, and do not read their reference files. The biggest saving here comes from a `PreToolUse` hook installed once, at zero context cost, forever; this text exists to set that up, and re-reading it afterwards costs more than it saves. Five pillars. Apply all by default; "verbose" or the user naming a model lifts the relevant rule. They engage by project condition, never all-or-nothing: no dev commands leaves RTK dormant, no subagents leaves routing idle. Pillars 2, 4 and 5 fit any project, code or not - SoT categories seed from the project goal in any domain (a writing project seeds BRIEF/VOICE/PLATFORM as a web app seeds AUTH/DB/API). Never force an inapplicable pillar. | Pillar | What it saves | Read | |---|---|---| | 1. RTK command proxy | 60-90% of command output | this file | | 2. Efficient file/command ops | wasteful reads and dumps | `references/strategies.md` | | 3. Model routing (hard rules) | expensive-tier tokens on cheap-tier work | `references/model-routing.md` | | 4. SoT keywords + hybrid memory | blind scanning and re-planning | `references/sot-framework.md`, `references/memory-system.md` | | 5. Concise output | narration and filler | `references/output-style.md` | ## Edition boundary - read before fol