Agent skill

isomorphic-hunter

Cross-domain isomorphic pattern forge. Dokkado Five Rings x meta-pattern 3+ domain validation, with arxiv evidence hunting and smithery tool discovery. RSI-refined. Finds what's ACTUALLY universal.

majiayu000github.com/majiayu000GitHub ↗
claude-codeMIT
Install
npx skills add majiayu000/claude-skill-registry --skill isomorphic-hunter --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 10 KB
Bundled scripts: none
Version: 1.0
Requires: - dokkado - meta-pattern-recognition - recursive-refiner - arxiv-search - gremlin-brain-v2
Path: skills/analysis/isomorphic-hunter/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Isomorphic Hunter **The skill that hunts what's real across domains.** Collision offspring: dokkado x meta-pattern-recognition, armed with arxiv search and smithery tool discovery, refined by recursive-refiner RSI loops. Not "this looks similar." IS this the same structure? Prove it or kill it. --- ## The Hunt Protocol ### Phase 0: ARM (Load Tools) Before hunting, check what weapons are available. ```bash # Search arxiv for prior art on the target pattern # (uses arxiv-mcp-server or direct API) arxiv_search() { local query="$1" # MCP server if available if command -v npx &>/dev/null; then echo "$query" | npx arxiv-mcp-server 2>/dev/null else # Direct API fallback curl -s "http://export.arxiv.org/api/query?search_query=all:$(echo "$query" | tr ' ' '+')" \ | python3 -c " import sys, xml.etree.ElementTree as ET root = ET.parse(sys.stdin).getroot() ns = {'a': 'http://www.w3.org/2005/Atom'} for entry in root.findall('a:entry', ns)[:5]: title = entry.find('a:title', ns).text.strip().replace('\n',' ') summary = entry.find('a:summary', ns).text.strip()[:200] link = entry.find('a:id', ns).text.strip() print(f'PAPER: {title}') print(f' URL: {link}') print(f' {summary}...') print() " 2>/dev

What's inside
Steps it walks through
  1. The Hunt Protocol
  2. Phase 0: ARM (Load Tools)
  3. Phase 1: GROUND (Dokkado Chi -- Extract Domain Morphemes)
  4. Phase 2: WATER (Dokkado Sui -- Discover Isomorphisms)
  5. Phase 3: FIRE (Dokkado Ka -- Derive the Kernel)
  6. Phase 4: WIND (Dokkado Fu -- Probing Attacks)
  7. Phase 5: VOID (Dokkado Ku -- Meta-Recursive Closure)
  8. Output Format
  9. Acceptance Criteria
  10. Failure Modes
  11. Apophenia
  12. Category Error
  13. Vacuous Universality
  14. Premature Closure
Ships with 1 file
  • metadata.json
Commands it runs
Search arxiv for prior art on the target pattern
local query="$1"
if command -v npx &>/dev/null; then
echo "$query" | npx arxiv-mcp-server 2>/dev/null
else
curl -s "http://export.arxiv.org/api/query?search_query=all:$(echo "$query" | tr ' ' '+')" \
import sys, xml.etree.ElementTree as ET
root = ET.parse(sys.stdin).getroot()
ns = {'a': 'http://www.w3.org/2005/Atom'}
for entry in root.findall('a:entry', ns)[:5]:
More from claude-skill-registry
All skills →
About this skill
What does the isomorphic-hunter skill do?

Cross-domain isomorphic pattern forge. Dokkado Five Rings x meta-pattern 3+ domain validation, with arxiv evidence hunting and smithery tool discovery. RSI-refined. Finds what's ACTUALLY universal.

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill isomorphic-hunter --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going