← 返回 Skills 市场

Clawdbot Logs Litiao

Name: Clawdbot Logs Litiao
Author: litiao1224

作者 litiao1224 · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

199

总下载

当前安装

版本数

在 OpenClaw 中安装

/install clawdbot-logs-litiao

功能描述

Analyze Clawdbot logs and diagnostics. Use when the user asks about bot performance, response times, errors, session stats, token usage, API costs, or wants...

使用说明 (SKILL.md)

Clawdbot Logs & Diagnostics

Analyze Clawdbot performance, errors, and session data.

Quick Commands

Response Times (last N messages)

scripts/response-times.sh [count]

Recent Errors

journalctl --user -u clawdbot-gateway.service --no-pager --since "1 hour ago" | grep -iE "(error|fail|invalid)" | tail -20

Session Stats

scripts/session-stats.sh

Gateway Status

systemctl --user status clawdbot-gateway.service --no-pager

Config Validation

cat ~/.clawdbot/clawdbot.json | jq . > /dev/null && echo "Config valid" || echo "Config invalid"

Log Sources

Source	Location	Contains
Journal	`journalctl --user -u clawdbot-gateway.service`	Session state, errors, tool exec
Daily log	`/tmp/clawdbot/clawdbot-YYYY-MM-DD.log`	Detailed JSON logs
Session file	`~/.clawdbot/agents/main/sessions/*.jsonl`	Full conversation, token usage, costs
Sessions meta	`~/.clawdbot/agents/main/sessions/sessions.json`	Current session state, model info

Common Diagnostics

Slow Responses

Check response times: scripts/response-times.sh 20
Check token count in sessions.json: jq '.["agent:main:main"].totalTokens' ~/.clawdbot/agents/main/sessions/sessions.json
If tokens > 30000, run /compact in Telegram or start new session

Config Errors

journalctl --user -u clawdbot-gateway.service --no-pager --since "10 minutes ago" | grep -i "invalid config"

API Costs (from session)

scripts/session-stats.sh

Useful Patterns

Filter journal by category

# Session state changes
journalctl --user -u clawdbot-gateway.service | grep "session state"

# Tool execution
journalctl --user -u clawdbot-gateway.service | grep "\[tools\]"

# Telegram activity
journalctl --user -u clawdbot-gateway.service | grep "\[telegram\]"

Parse session file for recent messages

tail -20 ~/.clawdbot/agents/main/sessions/*.jsonl | jq -r 'select(.message.role=="user") | .message.content[0].text' 2>/dev/null | tail -10

安全使用建议

This skill appears to do exactly what it claims: read Clawdbot journals and session files and summarize performance and cost data. Before installing or running it, remember: (1) it reads local session files (~/.clawdbot/...) that contain full conversation history, tokens, and cost information — treat that as highly sensitive; only run the skill on a trusted machine and avoid exposing its output to untrusted parties; (2) it relies on common system tools (journalctl, systemctl, jq, grep, tail) though they are not declared as required—ensure those binaries exist and are the expected versions; (3) there is a minor metadata mismatch (ownerId differs in _meta.json vs registry metadata) — likely benign but worth noting. If you want extra assurance, inspect the session files yourself before running or run the scripts with limited privileges / under a non-privileged account.

功能分析

Type: OpenClaw Skill Name: clawdbot-logs-litiao Version: 1.0.0 The skill bundle is designed for monitoring and debugging Clawdbot performance, errors, and session statistics. It uses bash scripts (scripts/response-times.sh and scripts/session-stats.sh) to parse local systemd journal logs and session data stored in ~/.clawdbot. The behavior is consistent with the stated purpose, and there are no indicators of data exfiltration, malicious command execution, or unauthorized persistence.

能力评估

✓ Purpose & Capability

Name/description align with requested actions: the SKILL.md and included scripts only read clawdbot journals, user systemd service status, and files under ~/.clawdbot to report response times, errors, token usage and costs. One minor mismatch: SKILL.md uses common system tools (journalctl, systemctl, jq, grep, tail) but the skill declares no required binaries—this is an omission but not a functional incoherence.

ℹ Instruction Scope

Instructions and scripts explicitly read sensitive local data (session JSON/JSONL containing full conversation history, token/cost info) and the user journal for the clawdbot service. All referenced paths and commands are within the expected scope for diagnostics; there are no steps that transmit data externally. Users should be aware that running this skill exposes local conversation contents to whoever or whatever is invoking the skill.

✓ Install Mechanism

Instruction-only skill with two small shell scripts included. No install spec, no downloads, and nothing is written to disk beyond running the provided scripts. This is low-risk from an installation perspective.

ℹ Credentials

The skill requests no environment variables or credentials, which is appropriate. However it reads files in $HOME and queries the user journal/systemd user service; access to these filesystem and journal entries grants visibility into sensitive conversation content and token/cost data—this is expected for a diagnostics tool but worth highlighting.

✓ Persistence & Privilege

always is false and there is no install-time privilege escalation or modification of other skills or system-wide config. The skill does not request persistent platform privileges beyond being invoked.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install clawdbot-logs-litiao
安装完成后，直接呼叫该 Skill 的名称或使用 /clawdbot-logs-litiao 触发
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

Bug fixes and improvements with -litiao suffix

元数据

Slug clawdbot-logs-litiao

版本 1.0.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 1

常见问题