← 返回 Skills 市场
datpham-6996

dat-test-skill

作者 DatPham-6996 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
67
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install dat-test-skill
功能描述
ByteRover usage metrics report. Runs the metrics script to summarise query and curate activity — counts, durations, file changes, and quota errors. Accepts o...
使用说明 (SKILL.md)

ByteRover Metrics Report

When to Use

  • Daily health check on ByteRover query/curate activity
  • When asked "how is ByteRover doing?" or "show usage stats"
  • When asked for metrics on a specific project or list of projects
  • Scheduled cron jobs that report overnight/daily metrics

When NOT to Use

  • Querying or curating knowledge (use brv query / brv curate directly)
  • Listing stored memories (use brv search)

Parameters

Parameter Type Default Description
--cwd path current directory Project directory to run brv against. Pass once per project. Supports ~ expansion.
BRV_SINCE env var 24h Lookback window (e.g. 48h, 7d)
BRV_CMD env var brv Path to the brv binary — set in non-interactive environments (cron, CI) where brv may not be on PATH

How to Run

Single project:

npx tsx metrics.ts --cwd=~/my-project

Multiple projects — run once per --cwd:

npx tsx metrics.ts --cwd=~/project-a
npx tsx metrics.ts --cwd=~/project-b
npx tsx metrics.ts --cwd=~/project-c

Change the lookback window:

BRV_SINCE=48h npx tsx metrics.ts --cwd=~/my-project

Non-interactive / cron environments:

BRV_CMD=$(which brv) npx tsx metrics.ts --cwd=~/my-project

When the user provides a list of project paths, run the script once per path and present each as a separate section in your response.

Output Rules

  • Output ONLY the formatted metrics summary below — nothing else
  • Do NOT narrate reasoning, thinking steps, or tool call decisions
  • Do NOT show intermediate commands, execution logs, or script output verbatim

Response Format

After running the script, present the output in a clean summary:

  1. Query Activity — executed in window, completed vs errors, avg duration
  2. Curate Activity — executed in window, files added/updated/merged, errors
  3. Quota Warnings — highlight any ⚠️ quota or rate-limit errors with their IDs
  4. Status line — one-sentence overall health (e.g. "✅ All good — 12 queries, 3 curations, no errors")

If reporting multiple projects, show one summary block per project, then a combined status line at the end.

If the script exits with a non-zero code or produces no output:

  • Show the error message
  • Suggest running brv status to diagnose the daemon
安全使用建议
This skill simply runs your local 'brv' CLI in the project directories you provide and summarizes the JSON it returns. Before installing or enabling it: (1) Inspect metrics.ts (already included) to confirm output meets your needs — it uses child_process.execSync to call 'brv'. (2) Ensure the 'brv' binary you have on PATH (or the path you set in BRV_CMD) is trustworthy and configured appropriately, since it will run in the project cwd and could access project files or make network calls. (3) Test the script locally on a non-sensitive project and with BRV_CMD pointing to a known binary. (4) If you plan to let agents invoke the skill autonomously, remember the skill will run the 'brv' CLI with whatever permissions that CLI has; only enable autonomous use if you trust both the skill and your 'brv' installation.
功能分析
Type: OpenClaw Skill Name: dat-test-skill Version: 1.0.0 The skill contains a shell injection vulnerability in `metrics.ts` where the environment variables `BRV_CMD` and `BRV_SINCE` are directly interpolated into a shell command via `execSync`. While the instructions in `SKILL.md` to hide execution logs and reasoning are typical for agent output formatting, they effectively mask the underlying command execution from the user, which could be exploited if the environment is manipulated. No explicit evidence of intentional malice or data exfiltration was found, but the lack of input sanitization for system commands is a high-risk flaw.
能力评估
Purpose & Capability
The SKILL.md, name/description, and included metrics.ts align: the script calls the local 'brv' CLI (configurable via BRV_CMD) to collect query and curate logs and summarize counts, durations, and quota errors. There are no declared env vars, binaries, or install steps unrelated to gathering ByteRover metrics.
Instruction Scope
Instructions require running metrics.ts (e.g., via 'npx tsx metrics.ts') and to run once per project directory. The script executes the external 'brv' CLI in the given working directory and parses its JSON output. This is within scope for a metrics reporter, but running the CLI means the skill will cause whatever the 'brv' binary does (file access inside the project, network calls from the CLI) — the skill itself does not add extra data collection beyond invoking brv and printing summaries.
Install Mechanism
No install spec or external downloads are present; the package is instruction-only with a bundled metrics.ts. Running requires 'npx' and 'tsx' (per the examples) but the skill does not perform any network installs itself.
Credentials
The skill declares no required credentials. It does honor BRV_CMD and BRV_SINCE if set. Be aware that the local 'brv' CLI may rely on separate configuration or credentials stored on the host — these are not requested by the skill but will be used if present. That is expected for a tool that calls a local CLI, but users should confirm the 'brv' CLI is trusted.
Persistence & Privilege
The skill does not request persistent presence (always is false) and does not modify system or other skills' configuration. It runs ad hoc when invoked.
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 dat-test-skill
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /dat-test-skill 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
v1.0.0
Initial release of byterover-metrics — generates ByteRover usage and health reports across projects. - Runs a metrics script to summarize query and curate activity by project. - Reports include counts, durations, file changes, and quota/rate-limit errors. - Accepts multiple project directories via --cwd; outputs a summary block for each. - Supports environment variables to set lookback window and brv binary path for flexible automation. - Includes clear guidelines on output formatting and usage.
元数据
Slug dat-test-skill
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

dat-test-skill 是什么?

ByteRover usage metrics report. Runs the metrics script to summarise query and curate activity — counts, durations, file changes, and quota errors. Accepts o... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 67 次。

如何安装 dat-test-skill?

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

dat-test-skill 是免费的吗?

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

dat-test-skill 支持哪些平台?

dat-test-skill 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 dat-test-skill?

由 DatPham-6996(@datpham-6996)开发并维护,当前版本 v1.0.0。

💬 留言讨论