Mindgraph

Create and query an Obsidian-style [[wikilink]] knowledge graph with learnable repeatable MindSkills for managing and analyzing workspace files.

Security Scan

The skill's code and instructions are internally consistent with its claim to index markdown files, build an Obsidian-style wikilink graph, and manage learnable 'MindSkills'; it reads and writes workspace markdown and a graph file but does not request credentials or reach out to external endpoints.

Purpose & Capability

The name/description (wikilinks + MindSkills) match the included SKILL.md and scripts: the Python script indexes .md files, resolves [[wikilinks]], and stores mindskill results under skills/mindgraph/mindskills/. No unrelated binaries, cloud credentials, or surprising resources are required.

Instruction Scope

SKILL.md explicitly instructs the agent to read workspace markdown files and to create/write MindSkill result files and the graph index (mindgraph.json). That behavior is consistent with the purpose, but the 'Always-On Rules' (use [[wikilinks]] for every markdown file you write) is a broad stylistic mandate that could affect many files and authoring workflows — review whether you want the agent to edit or standardize existing workspace files.

Install Mechanism

No install spec; this is an instruction-only skill that bundles a plain Python script. Nothing is downloaded from external URLs and there are no package-install steps in the manifest.

Credentials

No credentials or special environment variables are required. The script supports an optional MINDGRAPH_WORKSPACE env var to override the workspace root — this is reasonable and proportional.

Persistence & Privilege

always:false (normal). The skill does create persistent artifacts in the workspace (mindgraph.json, mindskill directories/results). That's expected for a knowledge-graph skill, but be aware these files are written into your workspace and persist across sessions.

Assessment

This skill appears to do what it says: it will scan all .md files in the workspace, write a mindgraph.json index in the workspace root, and create mindskill folders/results under skills/mindgraph/. It does not require credentials or contact external services. Before installing, consider: (1) If your workspace includes sensitive content in markdown, the index will read it — move or exclude sensitive files if needed. (2) The skill will persistently write mindgraph.json and result files; back up any existing files with the same names. (3) The SKILL.md enforces a broad authoring rule (always use [[wikilinks]]), which may change how the agent edits or writes notes — decide whether you want that convention applied automatically. (4) If you enable autonomous agent invocation, remember it could create/modify files when asked; consider limiting write actions to manual approval if you want stricter control.

A knowledge graph + learnable process engine for OpenClaw. Two core features:

1. Knowledge Graph — Obsidian-style [[wikilinks]] across all workspace files
2. MindSkills — Learned, repeatable processes that save graph-connected results

Always-On Rules

Every time you write or edit a markdown file, use [[wikilinks]] for:

• People: [[Alice]], [[Bob]]
• Projects: [[my-saas]], [[landing-page]]
• Companies/tools: [[Stripe]], [[Vercel]], [[GitHub]]
• Concepts/frameworks: [[Knockout Test]], [[B2B SaaS]]
• Other agents/models: [[Claude Code]], [[Sonnet]]

This is not optional. Links are how knowledge connects. No links = isolated notes = useless.

NEVER use [[wikilinks]] in messages to the user (Telegram, Discord, etc.). Wikilinks are for workspace files only. In conversations, write names plain: "Alice", not "[[Alice]]".

After significant file changes, rebuild the index:

python3 skills/mindgraph/scripts/mindgraph.py index

Graph Commands

# Build/rebuild index
python3 skills/mindgraph/scripts/mindgraph.py index

# Query a topic (backlinks + context + connections)
python3 skills/mindgraph/scripts/mindgraph.py query "<name>"

# Backlinks only (what references this?)
python3 skills/mindgraph/scripts/mindgraph.py backlinks "<name>"

# Forward links (what does this link to?)
python3 skills/mindgraph/scripts/mindgraph.py links "<file>"

# Bidirectional connections
python3 skills/mindgraph/scripts/mindgraph.py connections "<name>"

# ASCII tree visualization
python3 skills/mindgraph/scripts/mindgraph.py tree "<name>" [depth]

# Find orphans, dead links, unconnected files
python3 skills/mindgraph/scripts/mindgraph.py orphans
python3 skills/mindgraph/scripts/mindgraph.py deadlinks
python3 skills/mindgraph/scripts/mindgraph.py lonely

# Full statistics
python3 skills/mindgraph/scripts/mindgraph.py stats

MindSkills — Learned Processes

MindSkills are repeatable frameworks stored in skills/mindgraph/mindskills/. Each has a defined process and saves results as graph-connected markdown.

Using a MindSkill

# List all learned mindskills
python3 skills/mindgraph/scripts/mindgraph.py skills

# Show a mindskill's process
python3 skills/mindgraph/scripts/mindgraph.py skill <name>

# List results for a mindskill
python3 skills/mindgraph/scripts/mindgraph.py results <name>

When a user asks to run a process (e.g., "run the knockout test on X"), follow this flow:

1. Read the mindskill's PROCESS.md for the process definition
2. Execute the process conversationally
3. Save the result to skills/mindgraph/mindskills/<name>/results/<subject>.md
4. Use [[wikilinks]] throughout the result file
5. Include YAML frontmatter with metadata
6. Rebuild the graph index

Result file template:

---
mindskill: <skill-name>
subject: <what was tested/analyzed>
date: <YYYY-MM-DD>
verdict: <outcome>
aliases: [<aliases>]
---
# [[<MindSkill Name>]]: [[<Subject>]]

<Results following the process defined in PROCESS.md>

## Connections
- Related: [[link1]], [[link2]]

Learning a New MindSkill

When a user says "learn a mindskill called X" or describes a repeatable process:

# Create a new mindskill
python3 skills/mindgraph/scripts/mindgraph.py learn "<name>"

This creates the directory structure. Then write the PROCESS.md based on the user's description.

A good PROCESS.md contains:

• Purpose: What this process does and when to use it
• Trigger phrases: What the user might say to invoke this
• Steps: The actual process to follow (numbered)
• Output format: What the result file should contain
• Verdict/scoring: How to summarize the outcome (if applicable)

Discovering MindSkills

When a user's request matches a learned mindskill, proactively suggest it:

• "Want me to run the [[Knockout Test]] on that?"
• "I have an [[SEO Validator]] mindskill — should I audit that?"
• "This looks like a [[Competitor Analysis]] — want the full framework?"

Link Resolution

Links match (case-insensitive) against:

1. File basenames: [[MEMORY]] → MEMORY.md
2. Project dirs: [[my-saas]] → projects/my-saas/
3. MindSkill results: [[Pet Tracker KT]] → knockout test result
4. YAML aliases: aliases: [AV-Check] → [[AV-Check]] resolves
5. Unresolved → concept node (still tracked for backlinks)

File Locations

• Graph index: mindgraph.json (workspace root)
• MindSkills: skills/mindgraph/mindskills/
• Script: skills/mindgraph/scripts/mindgraph.py