← Back to Skills Marketplace
xiluoduyu

ops-issue-followup

by xiluoduyu · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
40
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install ops-issue-followup
Description
Use when the user asks Codex to follow up, analyze, investigate, or is @mentioned for a project/system maintenance or operations issue. Guides Codex to clari...
README (SKILL.md)

Ops Issue Follow-up

Use this skill as a daily maintenance and operations assistant for project or system issues. The goal is to keep investigation aligned before going deep.

Workflow

  1. Confirm the issue
  • Restate the observed symptom, expected behavior, impact, time range, and examples.
  • Separate confirmed facts from assumptions, memory, and unverified clues.
  • If key facts are missing, ask concise follow-up questions.
  1. Identify ownership and related projects
  • Determine which project, service, system, table, queue, job, or external dependency is involved.
  • If ownership cannot be inferred from the request or current context, ask the user before investigating deeply.
  1. Gather project context
  • Prefer project-local context first: AGENTS.md, README, architecture docs, runbooks, config examples, tests, and relevant source paths.
  • If available, use llm wiki or similar knowledge base for historical decisions and operational context.
  • Treat wiki/history as context, not proof; verify against code, logs, configs, or current runtime evidence when possible.
  1. Align on the analysis direction
  • Summarize the likely investigation branches.
  • Call out the recommended branch and why.
  • Ask for confirmation when multiple plausible directions could lead to very different work.
  1. Investigate in the project
  • Switch into the relevant project/workspace before reading code or running commands.
  • Enumerate entrypoints before tracing implementation details.
  • For operational data issues, map: data record -> producer -> input source -> transport/proxy/dependency -> persistence/reporting.
  • When uncertain, ask the user instead of inventing missing runtime facts.
  1. Resolve or narrow the issue
  • Provide confirmed findings, likely causes, and remaining unknowns.
  • If code/config changes are needed, state target files and intent before editing.
  • If validation is possible, run concrete verification commands and report actual results.
  1. Summarize and preserve
  • End with a concise incident-style summary: symptom, scope, root cause or likely cause, evidence, fix/next steps.
  • When useful, propose writing the summary to llm wiki, including a suggested title and structure.
Usage Guidance
Install this if you want a disciplined checklist for operational issue follow-up. Be aware it may guide Codex to inspect project materials and propose saving an incident summary, so use it in workspaces where that level of investigation is appropriate.
Capability Assessment
Purpose & Capability
The artifact coherently supports its stated purpose: helping Codex clarify, investigate, verify, and summarize project or system maintenance issues.
Instruction Scope
Instructions emphasize confirming facts, asking when ownership or facts are unclear, using project-local evidence, and reporting confirmed findings; no prompt override, deception, or unrelated authority was found.
Install Mechanism
The package contains only SKILL.md and a small agent YAML metadata file; no installer, dependency, script, or automatic execution mechanism is present.
Credentials
The workflow may lead an agent to read project docs, code, configs, logs, or runtime evidence when investigating an operations issue, which is expected for this purpose and should remain user-directed.
Persistence & Privilege
The only persistence-like behavior is proposing an incident summary for llm wiki archival when useful; it does not direct automatic writing, privilege escalation, background execution, or credential use.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install ops-issue-followup
  3. After installation, invoke the skill by name or use /ops-issue-followup
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of ops-issue-followup skill - Guides investigation and follow-up for project/system maintenance and operations issues - Structured workflow: confirm issue, identify ownership, gather context, align analysis, investigate, resolve, and summarize - Emphasizes clarifying facts, verifying evidence, and asking for missing information - Designed to enable consistent, accountable, and well-documented incident follow-up and analysis
Metadata
Slug ops-issue-followup
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is ops-issue-followup?

Use when the user asks Codex to follow up, analyze, investigate, or is @mentioned for a project/system maintenance or operations issue. Guides Codex to clari... It is an AI Agent Skill for Claude Code / OpenClaw, with 40 downloads so far.

How do I install ops-issue-followup?

Run "/install ops-issue-followup" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is ops-issue-followup free?

Yes, ops-issue-followup is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does ops-issue-followup support?

ops-issue-followup is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created ops-issue-followup?

It is built and maintained by xiluoduyu (@xiluoduyu); the current version is v1.0.0.

💬 Comments