New1Direction/korgex

The verifiable AI coding agent : provider-agnostic, MCP-native, tamper-evident ledger

This repository also configures its own agents. See what korgex tells them →

5Stars on the repository
53Mods indexed here, across every type
2mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

database-migrations

25

New1Direction/korgex

Skill Claude CodeCodex

Change a schema safely on a live system — reversible, ordered, zero-downtime.

not rated 5 2mo ago A 21 tokens original MIT

New1Direction/korgex

Skill Claude CodeCodex

Split independent work across parallel subagents with focused scope, then aggregate.

not rated 5 2mo ago A 19 tokens original MIT

deploying-safely

27

New1Direction/korgex

Skill Claude CodeCodex

Ship to production with staged rollout, health checks, and a fast rollback.

not rated 5 2mo ago A 20 tokens original MIT

error-handling

29

New1Direction/korgex

Skill Claude CodeCodex

Handle failures deliberately — fail loud, recover narrow, never swallow.

not rated 5 2mo ago A 16 tokens original MIT

New1Direction/korgex

Skill Claude CodeCodex

Map an unfamiliar codebase and find the right place to change before editing.

not rated 5 2mo ago A 20 tokens original MIT

handling-ambiguity

31

New1Direction/korgex

Skill Claude CodeCodex

When a request is unclear or underspecified, resolve it well instead of guessing.

not rated 5 2mo ago A 20 tokens original MIT

incident-response

33

New1Direction/korgex

Skill Claude CodeCodex

When production is broken — stop the bleeding first, diagnose second, learn third.

not rated 5 2mo ago A 18 tokens original MIT

issue-triage

34

New1Direction/korgex

Skill Claude CodeCodex

Turn a vague report into a clear, reproducible, prioritized issue.

not rated 5 2mo ago A 18 tokens original MIT

New1Direction/korgex

Skill Claude CodeCodex

Add, update, and audit third-party dependencies deliberately and safely.

not rated 5 2mo ago A 17 tokens original MIT

react-doctor

38

New1Direction/korgex

Skill Claude CodeCodex

Catch bad React in your changes — run react-doctor's deterministic scan and fix what it flags.

not rated 5 2mo ago A 23 tokens original MIT

safe-refactoring

40

New1Direction/korgex

Skill Claude CodeCodex

Change code structure without changing behaviour — tests green at every step.

not rated 5 2mo ago A 16 tokens original MIT

security-review

41

New1Direction/korgex

Skill Claude CodeCodex

Review code defensively for injection, secrets, and trust-boundary flaws.

not rated 5 2mo ago A 17 tokens original MIT

security-scan

42

New1Direction/korgex

Skill Claude CodeCodex

Run a verifiable security scan (vulns, secrets, misconfig) over code you wrote or deps you added, before calling the work done.

not rated 5 2mo ago A 34 tokens original MIT

spike

43

New1Direction/korgex

Skill Claude CodeCodex

Throwaway exploration to learn the unknown, then delete it and build it properly.

not rated 5 2mo ago A 18 tokens original MIT

New1Direction/korgex

Skill Claude CodeCodex

Reproduce, isolate the root cause, fix the cause not the symptom, then prove it.

not rated 5 2mo ago A 23 tokens original MIT

using-git

46

New1Direction/korgex

Skill Claude CodeCodex

Clean atomic commits, honest messages, and a safe branch/PR workflow.

not rated 5 2mo ago A 18 tokens original MIT

New1Direction/korgex

Skill Claude CodeCodex

Run multi-agent work as ONE provable causal DAG — parallel subagents, an Orchestrate graph, an immutable spec-seed.

not rated 5 2mo ago A 32 tokens original MIT

verify-before-done

48

New1Direction/korgex

Skill Claude CodeCodex

Prove the work is actually complete before claiming it — never report unverified.

not rated 5 2mo ago A 20 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: