用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/dvcrn/openclaw-skills-marketplace --skill reliability-evidence-pack命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
监控 OpenClaw GitHub 版本更新,获取最新版本发布说明,翻译成中文, 并推送到 Telegram 和 Feishu。用于:(1) 定时检查版本更新 (2) 推送版本更新通知 (3) 生成中文版发布说明
The philosophical layer for AI agents. Maps behavior to Spinoza's 48 affects, calculates persistence scores, and generates geometric self-reports. Give your agent a soul.
Order food/drinks (点餐) on an Android device paired as an OpenClaw node. Uses in-app menu and cart; add goods, view cart, submit order (demo, no real payment).
基于 SOC 职业分类
正在显示 SKILL.md
| name | reliability-evidence-pack |
| description | Reliability Evidence Pack |
A comprehensive runtime system for documenting and verifying agent operational reliability through structured artifact recording, validation, and compliance reporting.
/scripts)rep.mjs - Main CLI for initialization, validation, and bundle managementrep-validate.mjs - Schema validation engine for REP artifactsrep-heartbeat-cron.mjs - Records agent heartbeats on a schedulerep-near-miss-cron.mjs - Tracks near-miss reliability eventsrep-performance-baseline.mjs - Captures performance metricsrep-generate-bundle.mjs - Generates REP bundles from artifacts/cli)cli/README.md for installation and usage/github-action)github-action/README.md for setup/schemas)decision_rejection_log.jsonhandoff_acceptance_packet.jsonmemory_reconstruction_audit.jsonnear_miss_reliability_trailer.jsonsigned_divergence_violation_record.json/examples)No other system binaries required.
# Clone or copy this bundle to your project
cp -r rep-bundle-v2 /path/to/your/project/rep
cd rep
# Make scripts executable
chmod +x scripts/*.mjs
# Test
node scripts/rep.mjs --help
cd cli
npm install -g
rep --help
- uses: ./github-action
with:
bundle-path: ./rep
Set environment variables to configure behavior:
# Path to artifacts directory (default: ./artifacts)
REP_ARTIFACTS_PATH=./artifacts
# Path to schemas directory (default: ./schemas)
REP_SCHEMAS_PATH=./schemas
# Log file path (optional)
REP_LOG_FILE=/var/log/rep.log
node scripts/rep.mjs init --name "my-agent"
node scripts/rep-validate.mjs ./artifacts --strict
REP_ARTIFACTS_PATH=./artifacts node scripts/rep-heartbeat-cron.mjs
# Add to crontab - run heartbeat every 5 minutes
*/5 * * * * cd /path/to/rep && REP_ARTIFACTS_PATH=./artifacts node scripts/rep-heartbeat-cron.mjs >> /var/log/rep-heartbeat.log 2>&1
| Artifact | Purpose |
|---|---|
agent_heartbeat_record | Agent lifecycle events |
decision_rejection_log | Decisions and their outcomes |
context_snapshot | Memory/context state |
handoff_acceptance_packet | Inter-agent handoff validation |
near_miss_reliability_trailer | Near-miss events |
memory_reconstruction_audit | Memory integrity checks |
signed_divergence_violation_record | Policy violations |
MIT
SPEC.md, QUICKSTART.md, INTEGRATION.mdexamples/node scripts/rep-validate.mjs --helpREP captures context snapshots, decision logs, and memory-like artifacts that may contain sensitive information.
REP_ARTIFACTS_PATH to an isolated, access-controlled directoryThe SPEC includes signature fields for artifact integrity, but key management is operator-defined:
The GitHub Action is local to your repository:
github-action/entrypoint.sh before use in public CIAll scripts operate locally:
.gitignore