Knowledge base capture
Skill ClarentCinematics/Codex-Skills-for-Enterprise/skills/knowledge-base-capture
Convert meetings, support threads, incident learnings, project notes, SOP drafts, process explanations, or expert knowledge into clean knowledge-base articles with audience, steps, ownership, freshness metadata, related links, and review needs. Use when Codex needs to preserve institutional knowledge in a reusable article.From its SKILL.md
npx -y skills add ClarentCinematics/Codex-Skills-for-Enterprise --skill knowledge-base-captureAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 1 command, including `scripts/check_kb_metadata.py --input <draft.md>`.
SKILL.md
2.3 KB, 399 tokens by cl100k_base, as published. Nobody here has run it
Knowledge Base Capture
Workflow
- Identify audience, knowledge type, source material, owner, and intended knowledge base location if provided.
- Extract reusable knowledge: purpose, prerequisites, steps, decisions, troubleshooting, examples, and limits.
- Remove transient discussion, duplicated context, and unresolved claims unless they are marked for review.
- Structure the article for reuse, searchability, ownership, and future maintenance.
- Produce a clean KB draft with metadata and review questions.
Script-Assisted Workflow
When reviewing a markdown KB draft, run scripts/check_kb_metadata.py --input <draft.md> to check required sections, metadata completeness, review freshness, owner gaps, and unresolved assumptions. Adjust --required-sections and --today for the target knowledge-base standard. Use the output to decide publishing readiness; do not use the script to rewrite the article.
Output Standard
Use this structure by default:
- Title: searchable, specific, and outcome-oriented.
- Audience: who should use the article.
- Summary: what the article helps the reader do.
- Prerequisites: access, tools, context, or permissions needed.
- Procedure / Guidance: steps, decision rules, or reference content.
- Troubleshooting / Exceptions: known issues, limits, and escalation path.
- Ownership And Freshness: owner, review date, source, and confidence.
- Related Links: source artifacts or related docs when provided.
- Review Questions: unresolved items before publishing.
Rules
- Do not publish unresolved assumptions as facts.
- Preserve source provenance when available.
- Flag stale, sensitive, incomplete, or ownerless knowledge for review.
- Make the article useful to someone who was not present for the original discussion.
References
Read references/kb-patterns.md when the user asks for SOPs, troubleshooting articles, incident learnings, onboarding docs, or support knowledge.
What ships with it: 3 files
6.9 KB alongside SKILL.md, 1 of them executable
agents/
- openai.yaml230 B
references/
- kb-patterns.md936 B
scripts/
- check_kb_metadata.pyruns5.7 KB