#
ENJA

ClawOSS

billion-token-one-task/ClawOSS

ClawOSS is an autonomous open-source contributor agent framework designed to maximize merge rates for documentation fixes, typo corrections, test additions, and targeted bug fixes. It utilizes an orchestrator-subagent architecture with continuous web search, rigorous repo health checks, and strict conflict prevention.

3 14MITUpdated 2026-05-19

Overview

ClawOSS operates as an autonomous multi-agent system optimized for contributing to well-maintained open-source repositories. It employs a main orchestrator alongside persistent background subagents (such as scouts and PR monitors) and concurrent implementation agents. By prioritizing high-probability-of-merge contributions—like typo fixes, documentation updates, and small bug fixes—the system aims to build long-term reputation and trust across a curated list of repositories while rigorously avoiding duplicate work, CI breakage, and low-quality pull requests.

Capabilities

  • Autonomous issue discovery and triage using niche categorization
  • Multi-agent orchestrator architecture supporting concurrent implementation and review handling
  • Automated reproduction-first bug fixing with mandatory failing test verification
  • Pull request follow-up handling, review parsing, and iterative comment response
  • Repository health gate screening and metadata caching
  • Proactive web search for error research and upstream discussions

Best for

Discovering high-quality, recently active open-source issues across diverse technology stacks, Autonomously implementing, testing, and submitting small-scale bug fixes, documentation improvements, and typo corrections, Monitoring active pull requests, handling reviewer feedback, and managing follow-up iterations, Performing repository health checks and trust-scoring to target repositories with high merge rates

Works with

Claude CodeCursorCodex
Evoa Score breakdown= Σ (score × weight)
76
Task usefulness20%100 → +20.0

実タスクにどれだけ役立つか(機能の豊富さ・用途の明確さ)。 AIによるcapabilities/use-cases解析

Code quality15%90 → +13.5

実装・指示の品質。 AIによるSKILL.md/README解析

Maintenance15%85 → +12.8

リポジトリがどれだけ活発に保守されているか。 GitHub 最終push日時の新しさ

Documentation12%56 → +6.7

ドキュメントの充実度・分かりやすさ。 README/独自要約の情報量

Security15%50 → +7.5

危険・不審な挙動が無いか。 AIによるセキュリティレビュー

Originality10%85 → +8.5

ありふれたラッパーではない独自性。 AIによる独自性判定

Popularity8%38 → +3.0

コミュニティの採用度。 GitHub Stars/Forks(対数スケール)

Compatibility5%75 → +3.8

対応AIエージェントの広さ。 AIによる対応エージェント判定

Weighted total75.8 / 100

ライセンス不明/制限あり(Red)のSkillは総合スコアに0.85倍の補正を適用します。 ランキングはこのScoreのみで決まり、広告で変わりません。 算出方法の詳細 →

Security considerations

The skill enforces strict security boundaries including least-privilege GitHub token scopes (public_repo), prohibitions against pushing to master/main or force-pushing, and checks to prevent committing secrets, environment files, or PII.

Categories

Summary and analysis are original content generated by AI Skills Rank. The skill's source text is not reproduced here — view it on the linked repository.