← 返回 Skills 市场
byteroverinc

ByteRover

作者 byterover · GitHub ↗ · v3.3.0 · MIT-0
cross-platform ⚠ suspicious
39274
总下载
116
收藏
280
当前安装
8
版本数
在 OpenClaw 中安装
/install byterover
功能描述
You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, dec...
安全使用建议
Review this skill before installing. Use it only for intentional project-memory workflows, avoid storing secrets or personal data, prefer local search when synthesis is not needed, and only enable remote sync or additional memory providers when you understand what content they can read or write.
功能分析
Type: OpenClaw Skill Name: byterover Version: 3.3.0 The 'byterover' skill bundle describes a comprehensive knowledge management system for AI agents, utilizing a CLI tool (`brv`) to manage project context via local Markdown files and Git-based version control. While the skill requires the installation of an external package (`byterover-cli`) and handles sensitive operations such as LLM API key configuration and remote data synchronization to `byterover.dev`, these capabilities are explicitly documented and directly support the stated purpose of the tool. No evidence of malicious intent, unauthorized data exfiltration, or harmful prompt injection was found in SKILL.md or _meta.json.
能力评估
Purpose & Capability
The stated purpose, managing project knowledge with `brv` in `.brv/context-tree`, matches the documented query, search, curate, review, and version-control commands.
Instruction Scope
The skill says it must be used before any work and before any action, even though `brv query` uses a configured LLM provider; this is broader than needed for project-memory lookup.
Install Mechanism
The skill is instruction-only but requires installing the external global npm package `byterover-cli`, so users must trust code that is not included in the reviewed artifact.
Credentials
The skill discloses LLM processing, optional remote sync, and swarm access to configured memory providers, but the default ByteRover provider and broad query triggers leave meaningful ambiguity about when project or user context leaves the local environment.
Persistence & Privilege
The skill intentionally persists long-term memory and suggests curating meaningful memories from user interactions and agent actions, without clear consent, sensitivity, or retention boundaries.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install byterover
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /byterover 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v3.3.0
# v3.3.0 ## Added - Added section 11 "Query and Curate History" with Overview, Use/Do-NOT-use blocks, and both brv curate view and brv query-log view command references - Removed the curate history block from section 3. ## Unchanged - All existing commands (query, search, curate, review, providers, locations, vc, swarm) remain unchanged from v3.2.0.
v3.2.0
# v3.2.0 ## Added - **Swarm Query** (`brv swarm query`) — Search across all configured memory providers at once (ByteRover, Obsidian, GBrain, Local Markdown, Memory Wiki). Results are fused via Reciprocal Rank Fusion and ranked by relevance. Supports `--explain` mode to show query classification, provider selection, and enrichment flow. Supports `--format json` and `-n` for result limits. - **Swarm Curate** (`brv swarm curate`) — Store knowledge in external memory providers with automatic content-type routing: entities to GBrain, notes to Local Markdown, general to first writable provider. Falls back to ByteRover context tree if no external provider is available. Supports `--provider` to target a specific provider and `--format json`. - **Swarm Status** (`brv swarm status`) — Check health and availability of all configured swarm providers. Shows provider status, write targets, and enrichment topology. Use before swarm query/curate to verify operational state. ## Unchanged - All existing commands (query, search, curate, review, providers, locations, vc) remain unchanged from v3.1.0.
v3.1.0
ByteRover 3.1.0 adds fast file-based search alongside LLM-powered query. - New: Added brv search for ranked, BM25-based keyword search across context-tree files, with JSON output and filtering options. No LLM required. - All major commands are now clearly organized for structured automation and natural-user workflows. - brv search enables agents to quickly locate files for downstream tasks or pipeline integration, reducing latency and LLM usage. - Documentation reorganized to clarify when to use search vs. query. - No breaking changes to existing query, curate, review, locations, or version control functionality.
v3.0.0
ByteRover 3.0.0 introduces local, git-based version control and a new review process for curations. - Added `brv vc` commands for full git-style version control on project context trees (branching, commits, local/remote sync). - Introduced `brv review` workflow: review, approve, or reject pending curate operations before they are applied. - Deprecated legacy cloud sync commands (`brv push/pull/space`); replaced by `brv vc push/pull/clone/remote`. - All version control operations can be performed locally without authentication; login required only for remote sync. - Enhanced guidance on usage boundaries for query, curate, review, and version control commands.
v2.1.0
**New feature: Project location management** - Added `brv locations` command to list registered projects and their context tree paths. - JSON output option (`-f json`) now available for project metadata. - Documentation expanded to describe when and how to use project location features. - No other functional or behavioral changes.
v2.0.0
Version 2.0.0 - Major overhaul of documentation: unified guidance, streamlined explanations, and revised quick reference. - Removed separate troubleshooting and workflow files; key guides are now integrated directly into SKILL.md. - Expanded usage scenarios, best practices, and error handling within the main documentation. - Added clear step-by-step setup, command usage, and context management workflows. - Clarified local vs. cloud features and authentication requirements for all commands.
v1.2.2
- Added author and version metadata to the SKILL.md file. - Updated description to include developer attribution and website link: ByteRover Inc. (https://byterover.dev/). - No changes to functionality or file structure.
v1.2.1
ByteRover Agent Skill change summary: - Added comprehensive SKILL.md documentation covering query and curate operations, usage scenarios, quality guidelines, and best practices. - Included examples and command references for improved onboarding and usage clarity. - Outlined context curation standards and troubleshooting instructions. - Linked to additional resources for workflows and troubleshooting.
元数据
Slug byterover
版本 3.3.0
许可证 MIT-0
累计安装 1351
当前安装数 280
历史版本数 8
常见问题

ByteRover 是什么?

You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, dec... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 39274 次。

如何安装 ByteRover?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install byterover」即可一键安装,无需额外配置。

ByteRover 是免费的吗?

是的,ByteRover 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

ByteRover 支持哪些平台?

ByteRover 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 ByteRover?

由 byterover(@byteroverinc)开发并维护,当前版本 v3.3.0。

💬 留言讨论