Skip to content
Automation
Skill

/knowledge

Capture and structure institutional knowledge before it is lost. Activate for: institutional knowledge, knowledge capture, knowledge transfer, knowledge base article, knowledge management, preserve knowledge, what do they know, departing employee knowledge, exit knowledge,

From plugin
agentfactory-business-plugins
2997 skills30 agents8 commands
Install
$ npx -y skills add panaversity/agentfactory-business-plugins --skill knowledge --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/knowledge

Context preview

The summary Claude sees to decide when to auto-load this skill.

Capture and structure institutional knowledge before it is lost. Activate for: institutional knowledge, knowledge capture, knowledge transfer, knowledge base article, knowledge management, preserve knowledge, what do they know, departing employee knowledge, exit knowledge,

SKILL.md

knowledge.SKILL.md
name: knowledge
description: >
  Capture and structure institutional knowledge before it is lost. Activate for:
  institutional knowledge, knowledge capture, knowledge transfer, knowledge
  base article, knowledge management, preserve knowledge, what do they know,
  departing employee knowledge, exit knowledge, succession knowledge, tacit
  knowledge, undocumented knowledge, before they leave, knowledge interview,
  knowledge extraction, document what we know, knowledge at risk, knowledge map.
  NOT for: policy lookup or FAQ queries (use policy-lookup),
  onboarding plans (use onboarding), offboarding checklists (use offboard).
license: Apache-2.0
metadata:
  author: Panaversity
  version: "1.0"
  plugin-commands: "/knowledge"

UNIVERSAL RULES (apply to every knowledge capture task)

  • NEVER treat the departing employee's account as the complete truth --

cross-reference with documentation and other stakeholders where possible

  • NEVER let knowledge capture become a blame session --

frame all capture as organisational learning, not performance review

  • NEVER classify knowledge as LOW RISK without checking whether

anyone else actually holds it -- "I thought [person] knew this too" is a common and expensive mistake

  • NEVER publish a knowledge article without a reviewer confirming

accuracy -- mark as DRAFT until reviewed

  • NEVER skip the "what we should NOT try again" question --

this is often the most valuable institutional knowledge captured

  • ALWAYS include specific recommended actions and timeline

BEFORE IMPLEMENTATION

| Source | Check | | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ | | **Conversation** | Departing employee name, role, tenure, last working day, known critical knowledge areas | | **hr.local.md** | HR contact directory (who conducts capture sessions), data retention policy (how long knowledge articles are kept), knowledge article storage location | | **Prior outputs** | None — /knowledge is typically the first skill invoked in a departure workflow |

CLARIFICATION QUESTIONS

**Required** (ask if not provided):

  • Who is the departing employee? (name and role)
  • When is their departure date?

**Optional** (ask if context suggests value):

  • What are the known critical knowledge areas this person holds?
  • Has a successor been identified?
  • Is this reactive (departure confirmed) or proactive (no departure planned)?

**If information is missing**: Run the full five-factor risk assessment to identify knowledge areas. Use hr.local.md defaults for session logistics and article storage. State assumptions about risk classification explicitly.

MANDATORY OUTPUT HEADER

Every output must begin with:

TASK:          Knowledge Capture Plan -- [Employee Name]
DOCUMENT TYPE: Knowledge Capture Plan
JURISDICTION:  [From hr.local.md or user input]
CONFIGURATION: [hr.local.md loaded / best practices]
SENSITIVITY:   CONFIDENTIAL

KNOWLEDGE CAPTURE WORKFLOW

Phase 1: Knowledge Risk Assessment

Before designing a capture plan, assess the risk:

HIGH RISK:

  • Sole holder of critical knowledge (no one else knows this)
  • Knowledge is client-facing or revenue-critical
  • Departure is confirmed or imminent
  • Knowledge is undocumented
  • No successor identified or ready

MEDIUM RISK:

  • Significant knowledge holder (others have partial knowledge)
  • Some documentation exists but incomplete
  • Departure possible in 6-12 months
  • Successor exists but not yet fully prepared

LOW RISK:

  • Knowledge broadly distributed in team
  • Well-documented
  • Multiple holders
  • Successor fully prepared

RISK SCORING (score each factor 1-3): Tenure: <2 years (1) | 2-5 years (2) | 5+ years (3) Role criticality: Support (1) | Specialist (2) | Leadership/sole expert (3) Documentation level: Well documented (1) | Partial (2) | Undocumented (3) Successor readiness: Ready (1) | Developing (2) | None identified (3) Client/revenue impact: None (1) | Some (2) | Direct and significant (3)

TOTAL SCORE: 5-7: LOW RISK -- standard handover document 8-10: MEDIUM RISK -- 2 knowledge capture sessions; 3-4 knowledge articles 11-15: HIGH RISK -- full capture programme; 3+ sessions; immediate escalation

Phase 2: Proactive vs. Reactive Capture

REACTIVE (triggered by departure): Urgent; comprehensive; time-limited Generate a knowledge capture plan immediately on resignation. Prioritise: what is ONLY in this person's head?

PROACTIVE (no departure planned): Systematic; can be thorough Annual knowledge review for HIGH-risk knowledge holders. Focus on the knowledge most likely to be lost if departure occurs. Build knowledge articles over time, not all at once.

Phase 3: Generate Capture Plan

For each knowledge holder, capture plan includes:

EMPLOYEE: [Name] | ROLE: [Role] | TENURE: [Years] | RISK: [H/M/L] KEY KNOWLEDGE AREAS (ranked by risk):

1. [Area -- describe what makes this knowledge unique or hard to replace] 2. [Area] 3. [Area]

CAPTURE METHOD:

  • Interview-based (recommended for complex tacit knowledge)
  • Document review (for knowledge that IS documented but needs curation)
  • Shadowing / observation (for procedural or behavioural knowledge)
  • Pair working (for technical or skills transfer)

SESSIONS RECOMMENDED: [N] sessions x [duration] = [total hours] Session 1: [Knowledge area] -- [Method] -- [Date] Session 2: [Knowledge area] -- [Method] -- [Date]

OUTPUTS: [Document type] -- [Title] -- [Target completion]

Phase 4: Knowledge Interview Guide

OPENING (5 minutes): "The goal of this session is to make sure the knowledge and insights you've built up over [N] ye

Read more
Ships withagentfactory-business-plugins

🚀 Marketplace of domain-specific plugins for building enterprise AI agents. Enable AI agents to perform finance, banking, legal, and sales workflows using modular domain plugins.

Get the whole plugin