Governor – 토큰/컨텍스트 낭비를 줄이기 위한 Claude Code 플러그인
hackernews
|
|
📦 오픈소스
#claude
#command r
#eventsourcingdb
#mcp server
#이벤트 스토어
#오픈소스
원문 출처: hackernews · Genesis Park에서 요약 및 분석
요약
Governor는 Claude Code의 토큰 및 컨텍스트 낭비를 줄이기 위해 개발된 플러그인으로, 출력을 간결하게 유지하고 메모리 파일을 압축하며 불필요한 로그를 차단합니다. 이 도구는 단순한 토큰 절약을 넘어, 광범위한 작업에 대한 계획 가이드라인을 제공하여 효율성을 극대화합니다. 설치 후 플러그인이 로드되면 자동으로 컴팩트한 전문가 모드로 시작하며, 사용자는 명령어를 통해 기능을 켜고 끌 수 있습니다.
본문
Keep Claude Code concise, clean, and under control. Compact professional output, context hygiene, tool-output filtering, and usage telemetry for Claude Code Max users. Governor is the serious alternative to style-only token savers. It keeps the agent concise, shrinks recurring memory files, blocks noisy logs from flooding context, and adds planning guardrails for broad tasks. The installed Claude Code command namespace is still /governor:* . bash install.sh --force Restart Claude Code, then run: /governor:status /governor:audit /governor:compress CLAUDE.md Governor auto-starts in compact professional mode when the plugin is loaded. Use /governor:off to disable response compression and /governor:on to re-enable it. Heavy Claude Code users do not only burn quota on long answers. The bigger session killers are often: - bloated always-loaded context such as CLAUDE.md , notes, and rules - huge Bash/test/build output copied into conversation context - vague prompts that trigger broad scans and repeated failed attempts - scope drift during long coding tasks - compactions caused by preventable context growth Governor attacks those system problems while keeping the interaction professional and readable. Small local smoke benchmarks, not universal claims. Same machine, same prompt set, Claude CLI non-interactive runs. Three technical explanation prompts, Sonnet, no tools. | Condition | Output Tokens | Avg / Prompt | Saved vs Control | |---|---|---|---| | Control | 2967 | 989 | 0.0% | | Caveman | 1634 | 545 | 44.9% | | Governor | 1320 | 440 | 55.5% | One project-notes.md sample from the Caveman compression fixtures. | Method | Tokens | Saved | |---|---|---| | Original | 1877 | 0.0% | | Caveman fixture | 924 | 50.8% | | Governor medium | 838 | 55.4% | Synthetic noisy pytest -vv output with preserved failure lines. | Raw Output | Filtered Output | Blocked | |---|---|---| | 54314 estimated tokens | 1726 estimated tokens | 96.8% | Interpretation: Caveman is excellent at style compression. Governor aims for broader quota control: compact output, recurring-context compression, noisy-tool filtering, telemetry, and retry reduction. - Always-on compact mode: SessionStart andUserPromptSubmit hooks keep responses concise in every Claude Code chat. - Professional memory compression: /governor:compress CLAUDE.md rewrites verbose memory files into dense prose. - Protected-span safety: code blocks, inline code, paths, URLs, commands, env vars, versions, headings, tables, and warnings are preserved. - Quality guard: low-savings compression is rejected and the backup is restored instead of pretending success. - Tool-output filtering: noisy Bash/test/build output is summarized while preserving failure signals. - Telemetry ledger: /governor:status reports blocked tokens, failures, compactions, and statusline snapshots when available. - Prompt guidance: vague broad prompts get soft, non-blocking suggestions. - Plan and drift guard: explicit contracts for broad builds, then scope checks with /governor:guard . - Portable rule snippets: compact-mode rules for Codex, Gemini, Cursor, Windsurf, and Cline. | Command | Purpose | |---|---| /governor:on | Enable compact professional response mode | /governor:off | Disable response compression | /governor:status | Show usage dashboard and waste heat map | /governor:audit | Find bloated memory/rule files and context waste | /governor:compress CLAUDE.md | Compress memory files with protected-span validation | /governor:full | Let the next Bash command return full output | /governor:plan "task" | Produce an implementation contract before broad work | /governor:guard | Check current changes against the approved plan | /governor:benchmark | Explain or summarize benchmark results | /governor:install-rules | Copy compact-mode rules into other-agent projects | claude --plugin-dir . bash install.sh --force gh repo clone 0xhimanshu/governor cd governor bash install.sh --force bash install.sh --project /path/to/project --agents all Supported rule snippets: | Agent | Rule File | |---|---| | Codex | AGENTS.md | | Gemini CLI | GEMINI.md | | Cursor | .cursor/rules/governor.mdc | | Windsurf | .windsurf/rules/governor.md | | Cline | .clinerules/governor.md | Other agents get compact professional behavior only. Claude Code is the V1 target for hooks, telemetry, statusline, and Bash output filtering. /governor:compress is automatic from the user's point of view: - Create a timestamped backup. - Mark protected spans. - Rewrite the file with dense professional prose. - Strip markers. - Validate protected content. - Attempt protected-span recovery if needed. - Reject low-savings output and restore the backup if the quality guard fails. - Report exact before/after token estimates and backup location. Compression levels: | Level | Target | |---|---| light | Remove filler/repetition; preserve most rationale | medium | Collapse narrative into decision bullets | aggressive | Keep only rules, facts, commands, risks, and decisions | Set GOVERNOR_ALLOW_LOW_SAVINGS=1 only if you intentionally want to keep a low-savings compression result. Governor stores a local JSONL ledger under: ~/.claude/plugins/governor/ It tracks: - tool-output tokens blocked - full-output overrides - prompt-risk suggestions - Bash failures - compactions - statusline snapshots - memory compression savings Prompt caching can reduce usage/cost but does not necessarily reduce context window occupancy. Governor reports those separately when Claude Code exposes the data. Use benchmarks/ for measured comparisons. Recommended conditions: control : no token/style plugincaveman : Caveman enabled as normalgovernor-hooks : Governor hooks enabled, memory unchangedgovernor-compressed : Governor after/governor:compress CLAUDE.md governor-strict : optional strict-mode run for broad tasks Fill benchmarks/run-sheet.csv , then run: python3 scripts/compare_benchmarks.py benchmarks/run-sheet.csv Primary metrics: five-hour usage delta, peak context %, assistant output tokens, tool-output tokens blocked, failed tool calls, compactions, wall time, and task success. - Helpful by default, strict only when invoked. - Professional dense prose, never novelty dialect. - Measure exact savings where possible. - Treat 1M context as a ceiling, not a target. - Keep broad planning and drift checks opt-in. - Restore backups instead of accepting unsafe or low-value compression. - If hooks do not fire, /governor:status will show little or no telemetry. - Existing custom statuslines are not overwritten by the installer. - Compression sends file content through the active Claude Code/model workflow. Do not compress secrets or sensitive private files. - Use /governor:full before a diagnostic command when you need unfiltered logs. - For installed-but-inactive behavior, launch Claude Code with GOVERNOR_DEFAULT_MODE=off . Contributions are welcome when they make Governor more useful, safer, or easier to trust. Please read CONTRIBUTING.md before opening a pull request, especially for compression, hook, telemetry, or benchmark changes. MIT. See LICENSE.
Genesis Park 편집팀이 AI를 활용하여 작성한 분석입니다. 원문은 출처 링크를 통해 확인할 수 있습니다.
공유