consolidate
Connects scattered notes to improve knowledge graph density.
Install
mkdir -p .claude/skills/consolidate && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14744" && unzip -o skill.zip -d .claude/skills/consolidate && rm skill.zipInstalls to .claude/skills/consolidate
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Knowledge consolidation — connect orphans, suggest Maps, add wikilinksKey capabilities
- →Identify orphan notes with no inlinks
- →Find underconnected notes with few outlinks
- →Suggest new wikilink opportunities
- →Identify candidates for new knowledge maps
- →Detect possible note merges
- →Apply approved changes to wikilinks and maps
How it works
The skill scans notes to identify orphans and underconnected items, then suggests wikilinks, new maps, and merges based on content and existing links. It presents these as suggestions for user approval.
Inputs & outputs
When to use consolidate
- →Connect orphan notes
- →Create knowledge maps
- →Improve wikilink structure
About this skill
Time to consolidate the vault. Review scattered notes, find connection opportunities, and improve the knowledge graph.
Step 1: Scan orphans
Find notes that need attention by analyzing links:
- Read all Things: Use Glob to list all files in
Atoms/Things/. Read each one. - Check inlinks: For each Thing, search the rest of the vault (using Grep) to see if any other note links to it via
[[Note Name]]. Notes with zero inlinks are orphans. - Check outlinks: For each Thing, count how many
[[wikilinks]]it contains. Notes with fewer than 2 outlinks need more connections.
Step 2: Build connection map
Read all notes in Atoms/ and build a mental map of:
- What note links to what (outlinks via
[[wikilinks]]) - What note is linked by whom (inlinks)
- What topics they cover (via the
infrontmatter field)
Present a summary:
Vault health:
- Total Things: X
- Orphans (no inlinks): X
- Underconnected (fewer than 2 outlinks): X
- Well connected (≥1 inlink + ≥2 outlinks): X
- Sources: X
- People: X
- Maps: X
Step 3: Find wikilink opportunities
For each Thing that is an orphan or underconnected:
- Read its content
- Search in OTHER vault notes if they mention the same concept without linking
- Search if this note mentions concepts that exist as notes but aren't linked
Example suggestion:
Connection suggested:
In "Atoms/Things/Prompt Engineering.md" (line 5):
Before: "...language models need context..."
After: "...[[LLMs]] need context..."
Step 4: Identify new Map candidates
Review the in fields of all notes and find patterns:
- Are there topics in 5+ notes that don't have a Map?
- Are there notes without
inthat could belong to an existing Map?
Suggest new Maps only if there's critical mass (5+ notes).
Step 5: Detect possible merges
Look for notes that might be the same thing:
- Very similar names
- Significantly overlapping content
- Same topic from slightly different angles
Only suggest a merge if it's clear. When in doubt, don't.
Step 6: Present everything as suggestions
Show ALL suggestions organized by type BEFORE making any changes:
Consolidation suggestions:
Wikilinks to add (X):
1. In "Note A.md" (line N): "text" → "text with [[link]]"
New Maps to create (X):
1. "Maps/Topic.md" — groups: Note A, Note B, Note C...
Possible merges (X):
1. "Note A.md" and "Note B.md" — cover the same thing
`in` fields to add (X):
1. "Note A.md" → in: "[[Existing Map]]"
WAIT FOR APPROVAL — Don't execute anything until the user confirms which suggestions to apply.
Step 7: Execute approved changes
Only after confirmation:
- Wikilinks: Use Edit to add
[[links]]in note text. Links go integrated in text (Wikipedia-style), NEVER in "Related:" sections. - New Maps: Create following the existing Map pattern — with Dataview sections for Things, Sources, and People.
infield: Use Edit to update frontmatter with suggested Maps.
Step 8: Final summary
Show what was done:
Consolidation complete:
- X wikilinks added
- X Maps created
- X `in` fields updated
Important
- NEVER make changes without approval
- Links ALWAYS integrated in text, not in loose sections
When not to use it
- →When the knowledge vault is not structured with wikilinks and maps
- →When the user does not want to improve the knowledge graph
- →When the user does not want to connect scattered notes
Limitations
- →Changes are only executed after user approval
- →New maps are suggested only if there is critical mass (5+ notes)
- →Merges are suggested only if it is clear
How it compares
This skill automates the analysis of note connections and suggests improvements, unlike manually reviewing each note for linking opportunities.
Compared to similar skills
consolidate side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| consolidate (this skill) | 0 | 6mo | No flags | Intermediate |
| braindump | 7 | 5mo | No flags | Beginner |
| super-search | 10 | 2mo | Review | Beginner |
| mem-search | 8 | 4mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
braindump
huytieu
Quick capture of raw thoughts with intelligent domain classification and competitive intelligence extraction
super-search
supermemoryai
Search your coding memory. Use when user asks about past work, previous sessions, how something was implemented, what they worked on before, or wants to recall information from earlier sessions.
mem-search
thedotmack
Search claude-mem's persistent cross-session memory database. Use when user asks "did we already solve this?", "how did we do X last time?", or needs work from previous sessions.
baoyu-danger-x-to-markdown
JimLiu
Converts X (Twitter) tweets and articles to markdown with YAML front matter. Uses reverse-engineered API requiring user consent. Use when user mentions "X to markdown", "tweet to markdown", "save tweet", or provides x.com/twitter.com URLs for conversion.
session-search
pchalasani
For CLI agents WITHOUT subagent support (e.g., Codex CLI). Search previous code agent sessions for specific work, decisions, or code patterns.
hindsight-cloud
vectorize-io
Store team knowledge, project conventions, and learnings from tasks. Use to remember what works and recall context before new tasks. Connects to Hindsight Cloud. (user)