Add Skill Knowledge

VerifiedSafe

This skill captures new information learned during a session (troubleshooting, configurations, tools) and adds it to the appropriate skill knowledge base. It structures learnings with context, solution, and use cases for future reference. Helpful after solving a problem or discovering an important technique.

Sby Skills Guide Bot
DocumentationIntermediate
6/2/2026
Claude Code
#knowledge-capture#skill-documentation#learning-notes#claude-skills

Recommended for

Automated summary

This skill captures new knowledge learned during a session and adds it to the appropriate skill knowledge base for future reference.

Strengths

  • Builds a persistent knowledge base over time
  • Captures context and specific solutions to problems
  • Organizes learnings into clear categories (patterns, troubleshooting, etc.)
  • Integrates with end-of-day review for automatic capture

Limitations

  • Requires discipline to run regularly
  • May duplicate information if not checked against existing entries
  • Depends on the existing skill directory structure
When to use it

Use this skill after solving a problem or discovering a new configuration pattern you want to remember.

When not to use it

Do not use it for one-off fixes with no reuse potential or for information already well-documented elsewhere.

Security analysis

Safe
Quality score95/100

The skill provides instructions for capturing learned knowledge into markdown files, using git for version control. It does not involve any external command execution beyond file editing and git, and poses no risk of exfiltration, destruction, or safety bypass.

No concerns found

How this analysis is made

Examples

Capture a troubleshooting solution
/add-skill - I learned how to fix the PostgreSQL connection timeout by adjusting max_connections in the database skill.
Add a configuration pattern
We just discovered that setting broadcast storm control to 100 kpps prevents HA Pi network drops. Add this to the infrastructure skill under UniFi configuration.
End-of-day knowledge review
/eod - review the session for new learnings and add them to the appropriate skill files.

Add Skill Knowledge

Capture newly learned information and add it to the appropriate skill knowledge base.

Purpose

When you learn something new during a session (troubleshooting techniques, configuration patterns, tool usage), this command captures that knowledge and adds it to the relevant skill for future reference.


Usage

User invokes: /add-skill

Then specify:

  1. What was learned - Brief description
  2. Skill to update - Which skill this belongs to (infrastructure, scpi-automation, database, etc.)
  3. Category - What section to add it to (patterns, troubleshooting, configuration, tools, etc.)

Instructions

Step 1: Identify What Was Learned

Ask yourself:

  • What problem did we just solve?
  • What configuration/pattern did we discover?
  • What tool behavior did we learn?
  • What best practice emerged?

Step 2: Determine Target Skill

Available Skills:

  • infrastructure - Proxmox, Docker, networking (UniFi), storage
  • scpi-automation - Test equipment control and automation
  • database - PostgreSQL, homelab_db, ccpm_db operations
  • [future skills as needed]

Step 3: Add to Skill Knowledge

Read the skill's SKILL.md file and add the new knowledge to the appropriate section:

Common Sections:

  • Patterns - Reusable configuration patterns, templates
  • Troubleshooting - Solutions to specific problems
  • Configuration - Specific settings and their meanings
  • Tools - Tool usage patterns and best practices
  • Checklists - Pre/post task verification steps

Step 4: Format the Addition

Use this format:

### [Topic Name] (LEARNED: YYYY-MM-DD)

**Context:** [What problem/situation led to this learning]

**Solution/Pattern:**
```
[Configuration, command, or pattern]
```

**Why This Works:**
- [Explanation point 1]
- [Explanation point 2]

**When to Use:**
- [Use case 1]
- [Use case 2]

**Related:**
- [Links to docs, GitHub issues, session summaries]

Example Usage

User: /add-skill

You respond:

I'll add the knowledge we just gained. What did we learn?

Recent session context:
- HA Pi network failure troubleshooting
- UniFi storm control configuration
- STP/Loop Protection settings

Which should I capture?

User: "The UniFi port configuration for storm control and loop protection"

You:

Adding to: .claude/skills/infrastructure/SKILL.md
Section: UniFi Port Configuration
Category: Troubleshooting > Network Stability

[Shows the new section being added]

Auto-Capture During /eod

The /eod command now includes a step to review the session and identify learnings:

## Skill Knowledge Updates

During this session, we learned:
- [ ] UniFi storm control configuration (pKts/s thresholds)
- [ ] Loop Protection = STP Edge Port in UniFi
- [ ] Medical-critical network hardening patterns

Should I add these to skills? (Y/n)

Skill Knowledge Principles

What to Capture:

✅ Capture:

  • Configuration patterns that solved real problems
  • Tool behaviors discovered through troubleshooting
  • Best practices learned through experience
  • Settings and their actual effects (not documentation copy)
  • Troubleshooting procedures that worked

❌ Don't Capture:

  • One-off fixes specific to single issue
  • Information already well-documented elsewhere
  • Temporary workarounds
  • Obvious/common knowledge

Quality Over Quantity:

  • Be specific: "Storm control: Broadcast 100 kpps prevents HA Pi network drops"

  • Not generic: "Storm control prevents network problems"

  • Include context: "Loop Protection enabled on uplink ports prevents STP blocking during topology changes"

  • Not vague: "Enable loop protection"


Skill File Locations

.claude/skills/
├── infrastructure/
│   └── SKILL.md          # Proxmox, Docker, UniFi, networking
├── scpi-automation/
│   └── SKILL.md          # Test equipment patterns
├── database/
│   └── SKILL.md          # PostgreSQL, migrations, queries
└── [new-skill]/
    └── SKILL.md          # Created as needed

Creating New Skills

If knowledge doesn't fit existing skills:

  1. Create new skill directory: .claude/skills/[skill-name]/

  2. Create SKILL.md with structure:

    # [Skill Name] Skill
    
    **Skill Name:** [skill-name]
    **Version:** 1.0.0
    
    ## Purpose
    [What this skill covers]
    
    ## Allowed Tools
    [Tools this skill can use]
    
    ## Patterns
    [Learned patterns and configurations]
    
    ## Resources
    [Documentation references]
    
  3. Document in CLAUDE.md - Add to skills list


Verification

After adding skill knowledge:

  • [ ] Knowledge added to appropriate SKILL.md section
  • [ ] Includes context (what problem it solved)
  • [ ] Includes specific values/commands (not just concepts)
  • [ ] Dated with (LEARNED: YYYY-MM-DD)
  • [ ] Committed with descriptive message
  • [ ] Cross-referenced to session summary or GitHub issue

Git Commit Pattern

git add .claude/skills/[skill-name]/SKILL.md
git commit -m "skill: Add [topic] to [skill-name] knowledge base

Learned during [context/issue].

Added:
- [Specific pattern/config 1]
- [Specific pattern/config 2]

Solves: [Problem description]
Reference: [Session summary/issue link]
"

Example: Adding today's UniFi learning:

git add .claude/skills/infrastructure/SKILL.md
git commit -m "skill: Add UniFi storm control and STP configuration patterns

Learned during HA Pi network failure investigation (Issue #36).

Added:
- Storm control thresholds (pKts/s format)
- Loop Protection = STP Edge Port mapping
- Medical-critical network hardening patterns

Solves: Network drops during HA Pi service restarts
Reference: docs/session-summary-2026-01-10.md, Issue #36
"

HomeLab Agent - Continuous Learning System Version: 1.0

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/add-skill-knowledge-mm5t8bt8/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/add-skill-knowledge-mm5t8bt8' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/add-skill-knowledge-mm5t8bt8/download' --output '.skill-downloads/add-skill-knowledge-mm5t8bt8/SKILL.md.part' && mv '.skill-downloads/add-skill-knowledge-mm5t8bt8/SKILL.md.part' '.skill-downloads/add-skill-knowledge-mm5t8bt8/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/add-skill-knowledge-mm5t8bt8/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills