model-strategy

v0.4.1

Reduce total Claude Code and Codex usage by controlling context growth, turns, model effort, and bounded delegation. Strict routing audits are opt-in.

CodexClaude Code1 Skill

By 9uiLeLicense: MIT0 GitHub starsUpdated 1 hour ago

Directory evidence

Runtimes: Codex and Claude Code
Parsed components: 1 skill or MCP entry
Source updated: Oct 1, 2026
Manifest status: Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install model-strategy for Codex and Claude Code

Installs for the current user

codex plugin marketplace add 9uiLe/plugins
codex plugin marketplace upgrade 9uile-plugins
codex plugin add model-strategy@9uile-plugins

Paste and run these commands in a terminal with Codex. They add and refresh the 9uile-plugins catalog, then install this plugin.

Compatibility: the page URL and API slug “model-strategy” remain stable.

Codex: model-strategy@agent-plugin-marketplace → model-strategy@9uile-plugins

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually

$ git clone https://github.com/9uiLe/plugins

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/model-strategy/.

Plugin files

plugins/model-strategy/

├── .codex-plugin/plugin.json
├── .claude-plugin/plugin.json
└── skills/model-effort-guide/SKILL.md

Included Skills1

model-effort-guideskills/model-effort-guide/SKILL.md

Choose a cost-effective main model, effort, and bounded delegation plan for Claude Code or Codex when the user explicitly asks about model choice, usage cost, delegation, or operation routing. Do not invoke for routine coding merely because a task could be delegated. Japanese triggers: 「このタスクに最適なモデルは」「コスパよく実行して」「モデルと effort を選んで」「利用上限を節約して」「委譲方針を決めて」「操作をルーティングして」.

Plugin manifests2

plugins/model-strategy/.codex-plugin/plugin.json

{
  "name": "model-strategy",
  "version": "0.4.1",
  "description": "Reduce total Claude Code and Codex usage by controlling context growth, turns, model effort, and bounded delegation. Strict routing audits are opt-in.",
  "author": {
    "name": "9uiLe",
    "url": "https://github.com/9uiLe"
  },
  "homepage": "https://github.com/9uiLe/plugins/tree/master/plugins/model-strategy",
  "repository": "https://github.com/9uiLe/plugins",
  "license": "MIT",
  "keywords": [
    "model-selection",
    "effort",
    "cost-optimization",
    "subagents",
    "pricing",
    "token-economy"
  ],
  "skills": "./skills/",
  "interface": {
    "displayName": "Model Strategy",
    "shortDescription": "Control context, models, and delegation",
    "longDescription": "Reduce total usage with session boundaries, context hygiene, cost-effective model effort, and bounded delegation. Use strict routing and judge workflows only when explicitly requested.",
    "developerName": "9uiLe",
    "category": "Developer Tools",
    "capabilities": [
      "Read",
      "Write"
    ],
    "websiteURL": "https://github.com/9uiLe/plugins/tree/master/plugins/model-strategy",
    "defaultPrompt": [
      "Choose the most cost-effective model for this task",
      "Decide whether to delegate this implementation",
      "Explain the model and effort trade-off"
    ],
    "brandColor": "#7C3AED"
  }
}

plugins/model-strategy/.claude-plugin/plugin.json

{
  "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
  "name": "model-strategy",
  "version": "0.4.1",
  "description": "Reduce total Claude Code usage by controlling context growth and turn count before selecting model effort or bounded subagent delegation. Strict routing audits and conductor judges are explicit opt-ins.",
  "author": {
    "name": "9uiLe"
  },
  "homepage": "https://github.com/9uiLe/plugins",
  "repository": "https://github.com/9uiLe/plugins",
  "license": "MIT",
  "keywords": [
    "model-selection",
    "effort",
    "cost-optimization",
    "subagents",
    "pricing",
    "token-economy"
  ],
  "skills": "./skills/"
}

Browse the “model-selection” tag →

ai-coding-model-guidance

by Matt Corbett

Cross-tool AI-coding model guidance — three strategist agents that help you reason about model selection in the non-Claude coding ecosystems: GitHub Copilot's model picker (completions / chat / coding agent / cloud agent + org model rules), OpenAI Codex (CLI + cloud model lineup, reasoning levels, /model), and xAI Grok (Grok 4.x lineup + the coding-tool surface). Teaches the DURABLE reasoning — task-complexity → reasoning-level → cost-tier → surface — over a single citation-grounded, dated lineup that is the marketplace's freshness anchor for non-Claude model facts. Volatile specifics (prices, context windows, picker availability) carry retrieval dates + a re-verify-at-use rider, and the agents fetch live before quoting. Seam: when the answer is a CLAUDE model or a Claude-app build, escalate to claude-app-engineering/claude-solution-architect — this plugin owns the non-Claude tools. Requires ravenclaude-core@>=0.7.0.

Claude Code9 Skills

7Updated 2 weeks ago

gates

by Jason Matthew

Four stateless PreToolUse gates, a PostToolUse JSON-config guard, and a docs-consolidation trigger. Denies a git commit that changes code without staging its covering docs (per-plugin README/CLAUDE.md pairs, or the nearest README/CLAUDE/AGENTS walking up); asks before a new-project scaffold command (npm create vite, cargo new, rails new, …) runs ahead of an approved design; denies a high-fan-out Workflow script with no per-agent model: override and asks before a denylisted built-in named workflow; denies an ad-hoc Agent dispatch that omits model unless the agent type pins one in frontmatter. Separately, a never-blocking trigger counts commits since the .docs-sync record and nudges toward /docs-consolidate. Every gate has an ack marker. Separately, a PostToolUse hook re-parses settings.json/.mcp.json after an edit and reports a syntax error, which Claude Code otherwise swallows by silently dropping the config it cannot parse.

Claude Code1 Skill

5Updated 3 weeks ago

fugu-router

by yukineko

v0.1.28: fixes the defect v0.1.27's wiring fix exposed — the hook was reachable again, but `sync` could not finish from the store's own steady state. `cmd_sync` pulled BEFORE committing local appends, and pulled with `--ff-only`. Both abort in normal operation: the store files live inside the sync dir, so every `record` leaves them as uncommitted working-tree modifications (`git pull` then refuses with "Your local changes to the following files would be overwritten by merge", exit 1), and any second machine pushing makes the histories diverge (`--ff-only` then exits 128, "Not possible to fast-forward"). Both were reproduced against a copy of the real store. The order is now commit → pull `--no-rebase --no-edit` → push, with the push skipped only when the branch can be *established* to be level with its upstream (an undeterminable position pushes anyway). Reordering alone was still not enough: two machines appending to the same JSONL land their additions adjacent at end-of-file, which the default merge driver reports as a content conflict, so `sync` now ensures the sync dir's `.gitattributes` declares `episodes.jsonl`/`playbooks.jsonl` as `merge=union` — it keeps BOTH sides' lines, and the store is already deduplicated by content hash, so a duplicate line is recoverable where a dropped episode is not. Finally, a failed sync no longer fails into silence: its only caller is a `SessionEnd` hook whose exit code and stderr reach nobody, so the failure is written to `~/.fugu-router/sync-error.json` and surfaced by this plugin's own `UserPromptSubmit` hook (as `additionalContext` for the model AND `systemMessage` for the user) until a later sync succeeds and clears it. An unreadable or unparseable marker still produces a notice — its presence already says the last sync failed. New test `sync_ordering.rs` (4 tests, RED observed before GREEN) covers the steady state, idempotency on repeat, surfacing, and clearing. v0.1.27: `hooks/hooks.json` now declares the record-store sync as a `SessionEnd` hook (`${CLAUDE_PLUGIN_ROOT}/bin/fugu-router sync`, timeout 30). The wiring previously existed only as an orphaned absolute path in the user's `~/.claude/settings.json` (`~/.cargo/bin/fugu-router sync`), whose build had been renamed away — so the hook exited 127 every session with neither exit code nor stderr reaching the agent or the user, making a sync that had nothing to do indistinguishable from a sync that never ran (CLAUDE.md §1/§3: it went dark, not red). The damage was silent and cumulative — measured 2026-08-26: the last `fugu-router sync` commit in `~/.fugu-router/record-repo` is 2026-07-23 09:57:38 +0900, while `episodes.jsonl`/`playbooks.jsonl` in that repo kept growing as uncommitted, unpushed working-tree modifications for ~34 days, so every episode recorded in that window existed only on one machine. New test `session_end_sync_hook_wired.rs` pins the structural fix: the declaration must live in-plugin, every declared command must resolve through `${CLAUDE_PLUGIN_ROOT}` and never through `~/.cargo/bin`, and every declared subcommand must actually exist in the binary (RED observed before GREEN). fugu-style per-model routing for Claude Code orchestration. Learns from past task outcomes (which model passed verification, at what cost) and picks the cheapest Claude tier that historically clears similar work. Feeds condukt's suggested_model deterministically; records outcomes back to a local episode store. No API key, no embedding service — lexical k-NN over a JSONL store. v0.1.24: new mode axis `--mode fast|normal|high` for `route`/`suggest` — a deterministic clamp applied AFTER the policy has already picked a worker/verifier pair (never touches `decide`/`decide_bandit`'s learning logic). `fast` shifts the worker one tier down (capped at sonnet — opus never selected as worker or verifier); `high` shifts one tier up (capped at opus); `normal` is the backward-compatible identity default. The verifier is always recomputed from the clamped worker via `policy::verifier_model`, additionally capped at sonnet under `fast`. Precedence: `--mode` flag > env `FUGU_ROUTER_MODE` > config.toml `mode` > `normal`; an invalid env/config value is an explicit error + non-zero exit, never silently coerced to `normal`. A `gated` decision is a no-op under every mode (mirrors `downgrade_for_budget`'s own gated no-op). Ordering: the mode clamp runs BEFORE `downgrade_for_budget` — budget is a hard resource limit and wins over a mode preference, and the rationale records both the mode's shift and any budget negation so a `high` pick never silently disappears. `Episode` gains a measurement-only `mode: Option<String>` field (`record --mode`); an absent value means "not recorded", never conflated with `Some("normal")`. v0.1.20: new `duration-outliers` command flags models whose avg duration within a task class is a relative outlier vs other models in the same class (>1.5x the cross-model mean by default), and cross-references outlier-vs-normal effective pass rate — the measurement tool for PDO hypothesis ae64db03. v0.1.19: `Episode` gains measurement-only `route_basis`/`route_confidence`/`route_rationale` (the routing `Decision`'s provenance), `lines_added`/`lines_removed`, and `tokens_input`/`tokens_output`, all `Option<_>` and backward-compatible (older JSONL lines parse with these as `None`); `record` gains matching `--route-basis`/`--route-confidence`/`--route-rationale`/`--lines-added`/`--lines-removed`/`--tokens-input`/`--tokens-output` flags. None of this is consulted by `policy::route`/`decide_bandit` — it exists so routing decisions and task cost can be retrospectively correlated against actual pass/fail outcomes. v0.1.12: `store::append_playbook` now writes body+newline in one `write_all` call (mirroring `append`'s existing single-syscall pattern) instead of `writeln!`'s two syscalls, closing the same O_APPEND interleaving hazard for playbook records; a JSON-serialization failure now propagates as an `io::Result` error instead of silently writing an empty line. New regression test `concurrent_append_playbook_never_interleaves_records`. v0.1.13: every git subprocess in `cmd_sync` (pull/clone/status/add/diff/commit/push) is now bounded by a timeout (30s network ops, 10s local ops) via `wait-timeout`, killing a hung/stalled git process instead of wedging the sync command indefinitely. v0.1.14: `Episode` gains a measurement-only `duration_secs: f64` field (`#[serde(default)]`, backward-compatible with older JSONL lines) and `record` gains a `--duration` flag threading it through; not consulted by `policy::route`/`decide_bandit` — routing/scoring behavior is unchanged.

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[model-strategy on Agent Plugins Marketplace](https://pluginsmp.com/plugins/model-strategy)