name	code-standards
description	Dispatch independent adversarial reviews for ZFC, ZFC leveling, and root-cause-first without duplicating their standards.

Code Standards

Use this skill to dispatch code, diff, PR, or proposed-implementation review against the repo's core standards. This is a pointer/dispatch skill: reference the source skills below instead of duplicating their standards.

Source Skills

Load the relevant source skills by path and treat them as authoritative. These live at ~/.claude/skills/ (user-scope, shared across all repos) and are mirrored under .codex/skills/ as pointer files. Skill discovery resolves personal skills before project skills.

.claude/skills/zero-framework-cognition/SKILL.md (/zfc)
.claude/skills/zfc-leveling-roadmap/SKILL.md (/zfclevel)
.claude/skills/root-cause-first/SKILL.md (/root-cause-first)
Thermo-nuclear: invoke via Agent tool with subagent_type: thermo-nuclear-code-quality-review (/thermo)

Workflow

Define the scope from the command argument. If no argument is provided, review the current diff or active PR context.
Dispatch four adversarial, independent review lanes every time: /zfc, /zfclevel, /root-cause-first, and /thermo.
Give each lane only its scope, the relevant source skill path, and the instruction to look for blockers with file/line evidence. Do not paste or restate the full standard into the prompt.
For Claude callers, use a Codex subagent or Codex reviewer plugin for at least one independent lane when available, so one reviewer is not relying on the same model/context as the caller.
The /thermo lane uses subagent_type: thermo-nuclear-code-quality-review (NOT a bash command). Pass: diff summary, file list, working dir.
Require every lane to report PASS/WARN/FAIL, evidence, required fixes, and any N/A caveats. A lane may return N/A for a subsection only if the source skill permits that; the lane itself still runs.
Reconcile the lane results into one practical report. Do not dilute a FAIL from any required lane into a PASS.

Report Format

Return a concise report:

## Code Standards Report

Scope: <file, diff, PR, or task>

- ZFC: PASS/WARN/FAIL - <evidence>
- ZFC leveling: PASS/WARN/FAIL - <evidence or permitted N/A caveat>
- Root-cause-first: PASS/WARN/FAIL - <evidence or permitted N/A caveat>
- Thermo: PASS/WARN/FAIL - <evidence or structural blockers>

Blockers:
- <line-level issue and required fix>

Next checks:
- <tests, evidence, or review steps needed>

If there are no blockers, say so explicitly and list any residual risks. Do not mark any of the four lanes as skipped.

Mehr aus diesem Repository

gleiches Repository

adjustment-proof

jleechanorg/claude-commands

Use when proving, rejecting, or deleting backend adjustment registry entries with real LLM evidence and independent disabled-adjustment worktrees.

2026-06-0227

ao-operator-discipline

jleechanorg/claude-commands

Use when spawning, steering, or auditing Agent Orchestrator workers, especially when the user specifies exact AO parameters such as codex, runtime, project, claim-pr, or PR targets.

2026-06-0227

ao-worker-dispatch

jleechanorg/claude-commands

Checklist for dispatching AO workers — python venv, commit discipline, branch drift, and post-push CodeRabbit verification

2026-06-0227

claw-dispatch

jleechanorg/claude-commands

Use when dispatching work through the Hermes gateway with /claw, especially when the task may resolve slash commands or hand off into AO worker orchestration.

2026-06-0227

dark-factory

jleechanorg/claude-commands

Run the Dark Factory DOT pipeline runner against a goal. Slash command: /factory. Implements StrongDM's Attractor pattern as an external Python runner — .dot files are the versioned artifact, sealed holdouts live in a separate repo, every step is recorded to CXDB, and the Healer clusters failures into diagnoses. Use when you want the goal_harness idea executed as a reproducible external pipeline instead of in-Claude subagent dispatch.

2026-06-0227

disk-audit

jleechanorg/claude-commands

Run disk usage analysis and cleanup preview on the local Mac. Always validate snapshot coverage before quoting. Never delete without user approval.

2026-06-0227

Code Standards

Source Skills

.claude/skills/zero-framework-cognition/SKILL.md (/zfc)

.claude/skills/zfc-leveling-roadmap/SKILL.md (/zfclevel)

.claude/skills/root-cause-first/SKILL.md (/root-cause-first)

Thermo-nuclear: invoke via Agent tool with subagent_type: thermo-nuclear-code-quality-review (/thermo)

Workflow

Define the scope from the command argument. If no argument is provided, review the current diff or active PR context.

Dispatch four adversarial, independent review lanes every time: /zfc, /zfclevel, /root-cause-first, and /thermo.

Give each lane only its scope, the relevant source skill path, and the instruction to look for blockers with file/line evidence. Do not paste or restate the full standard into the prompt.

For Claude callers, use a Codex subagent or Codex reviewer plugin for at least one independent lane when available, so one reviewer is not relying on the same model/context as the caller.

The /thermo lane uses subagent_type: thermo-nuclear-code-quality-review (NOT a bash command). Pass: diff summary, file list, working dir.

Require every lane to report PASS/WARN/FAIL, evidence, required fixes, and any N/A caveats. A lane may return N/A for a subsection only if the source skill permits that; the lane itself still runs.

Reconcile the lane results into one practical report. Do not dilute a FAIL from any required lane into a PASS.

Report Format

Return a concise report:

## Code Standards Report Scope: <file, diff, PR, or task> - ZFC: PASS/WARN/FAIL - <evidence> - ZFC leveling: PASS/WARN/FAIL - <evidence or permitted N/A caveat> - Root-cause-first: PASS/WARN/FAIL - <evidence or permitted N/A caveat> - Thermo: PASS/WARN/FAIL - <evidence or structural blockers> Blockers: - <line-level issue and required fix> Next checks: - <tests, evidence, or review steps needed>

If there are no blockers, say so explicitly and list any residual risks. Do not mark any of the four lanes as skipped.