← Back to Skills Marketplace
lookupmark

LookupMark Log Analyzer

by LookUpMark · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ pending
114
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install lookupmark-log-analyzer
Description
Securely analyze system and application logs with automatic sensitive data redaction. Supports OpenClaw gateway logs (journalctl), RAG indexing logs, and que...
README (SKILL.md)

Log Analyzer

Secure log analysis with automatic redaction of tokens, keys, and passwords.

Usage

# Error summary across all sources (default)
python3 scripts/analyzer.py

# Specific source
python3 scripts/analyzer.py --source openclaw      # Gateway logs
python3 scripts/analyzer.py --source rag            # RAG indexing logs
python3 scripts/analyzer.py --source queries        # RAG query logs

# Search across all logs
python3 scripts/analyzer.py --search "OOM"

# Show only errors
python3 scripts/analyzer.py --source openclaw --errors

# More lines
python3 scripts/analyzer.py --source openclaw --last 500

Log Sources

Source Type Path
openclaw / gateway journalctl openclaw-gateway unit
rag file ~/.local/share/local-rag/index-batch.log
queries file ~/.local/share/local-rag/queries.log

Security

  • ALLOWED_SOURCES only: Cannot read arbitrary log files
  • Auto-redaction: Tokens, API keys, passwords, age keys are replaced with [REDACTED]
  • Read-only: Never modifies log files
  • No network: All local
  • Safe in groups: Sanitized output contains no secrets

Error Detection

Automatically detects: ERROR, FATAL, OOM, SIGKILL, FAILED, timeout, refused, denied, traceback, exception, segfault

Usage Guidance
This skill appears to be what it claims: a local, read-only log analyzer with redaction. Before installing or running it, review the included script and be aware of these points: (1) it reads only the journal unit openclaw-gateway and two files under ~/.local/share/local-rag — verify those are the logs you expect; (2) it loads ~/.config/log-analyzer/config.json which can change redaction or error patterns — inspect or create that file if you need stricter redaction; (3) the --search pattern accepts arbitrary regexes from the user — avoid running untrusted or extremely complex regexes (ReDoS); (4) redaction is regex-based and may not catch every secret — do not paste sanitized output into untrusted external services without spot-checking; (5) it invokes journalctl via subprocess (user scope), so ensure you run it as the intended user and not with elevated privileges. If you want extra assurance, run the script locally and inspect its output before granting it any broader automated invocation.
Capability Assessment
Purpose & Capability
Name/description claim secure log analysis with redaction; the included script implements that and only targets the declared sources (journalctl unit openclaw-gateway and two local RAG log files). No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md and the script are consistent: they limit reads to ALLOWED_SOURCES and state read-only/no-network. The script does load a user config (~/.config/log-analyzer/config.json) to override error and redact patterns — this can change detection/redaction behavior if the local config is modified. Also user-supplied search regexes are executed locally (could cause expensive regex evaluation).
Install Mechanism
No install spec and only a small Python script are provided, so nothing is downloaded or written to disk by an installer. This is the lowest-risk installation model.
Credentials
The skill requests no environment variables, no credentials, and no special config paths beyond a user-scoped config file for optional pattern overrides. Those config and log paths are proportionate to a log analyzer.
Persistence & Privilege
The skill is not always-enabled and does not modify other skills or system settings. It reads a per-user config file and local logs only; it does not create persistent system-wide privileges.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install lookupmark-log-analyzer
  3. After installation, invoke the skill by name or use /lookupmark-log-analyzer
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
Configurable error/redact patterns via config file. Added base64 and card number redaction.
v1.0.0
Initial release. journalctl + file log analysis. Auto-redaction of tokens/keys. Error detection and summarization.
Metadata
Slug lookupmark-log-analyzer
Version 1.1.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is LookupMark Log Analyzer?

Securely analyze system and application logs with automatic sensitive data redaction. Supports OpenClaw gateway logs (journalctl), RAG indexing logs, and que... It is an AI Agent Skill for Claude Code / OpenClaw, with 114 downloads so far.

How do I install LookupMark Log Analyzer?

Run "/install lookupmark-log-analyzer" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is LookupMark Log Analyzer free?

Yes, LookupMark Log Analyzer is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does LookupMark Log Analyzer support?

LookupMark Log Analyzer is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created LookupMark Log Analyzer?

It is built and maintained by LookUpMark (@lookupmark); the current version is v1.1.0.

💬 Comments