cubicloop

Iteratively improves local code changes by repeatedly running cubic CLI review, fixing actionable findings, and rerunning until no actionable issues remain (or an iteration cap is reached). Use when the user wants a local pre-flight quality loop before push/PR using cubic review on uncommitted changes, base-branch diffs, specific commits, or focused prompt-based checks.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "cubicloop" with this command: npx skills add notkainoa/cubic-review-loop/notkainoa-cubic-review-loop-cubicloop

Cubicloop

Iteratively run cubic review, fix actionable issues, and rerun until the review is clean.

Inputs

  • Review target:
    • Current branch against main: cubic review --base main
    • Uncommitted changes: cubic review
    • Specific commit: cubic review --commit <ref>
  • Custom focus (optional): cubic review --prompt "<instructions>"
  • Max iterations (optional): default 5

Workflow

1. Preflight

  1. Confirm cubic is available and authenticated.

  2. Confirm current workspace is a git repo.

  3. If the user did not specify a review target, ask a simple choice question:

    Do you want to:

    • review the current branch against main
    • review uncommitted changes
    • review a specific commit
    • focus on a specific area or problem
  4. Pick a single review mode based on the user's answer.

Use exactly one of:

  • --base main or another explicit base branch
  • uncommitted changes
  • --commit <ref>
  • --prompt "<instructions>"

--base, --commit, and --prompt are mutually exclusive. See CLI JSON reference.

2. Iteration Loop (Max 5)

For iteration N from 1 to max:

  1. Run review and store JSON output:
    mkdir -p .cubicloop
    cubic review --json > ".cubicloop/iteration-${N}.json"
    
    If using a mode, include it before --json (for example cubic review --base main --json).
  2. Parse issue totals and priority counts directly from the JSON output. jq is optional; use it only if available. See CLI JSON reference.
  3. Exit early if issue count is 0.
  4. Fix actionable issues in priority order: P0, P1, then the rest.
  5. For each issue:
    • Read the file and line context.
    • Decide if actionable or likely false positive.
    • Apply a fix for actionable items.
    • Record non-actionable items in the report with rationale.
  6. Run relevant validation (tests/lint/typecheck) when available.
  7. Continue to next iteration.

3. Exit Conditions

Stop when either condition is true:

  • No remaining actionable issues in Cubic output.
  • Max iterations reached.

4. Report

After exiting, summarize:

FieldValue
IterationsN
Target modeuncommitted / base / commit / prompt
Issues fixedN
Remaining issuesN
Highest remaining priorityP0/P1/P2/P3 or none

If max iterations are reached, include a short list of remaining issues with file and line.

Output Format

If fully resolved:

Cubicloop complete.
  Iterations:    2
  Target:        --base main
  Fixed:         6 issues
  Remaining:     0

If not fully resolved:

Cubicloop stopped after 5 iterations.
  Target:        uncommitted
  Fixed:         11 issues
  Remaining:     2
  Highest left:  P1

Remaining issues:
  - src/auth.ts:45 [P1] SQL injection vulnerability in user lookup
  - src/db.ts:112 [P2] Missing index on user_id column

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

openclaw-version-monitor

监控 OpenClaw GitHub 版本更新,获取最新版本发布说明,翻译成中文, 并推送到 Telegram 和 Feishu。用于:(1) 定时检查版本更新 (2) 推送版本更新通知 (3) 生成中文版发布说明

Archived SourceRecently Updated
Coding

ask-claude

Delegate a task to Claude Code CLI and immediately report the result back in chat. Supports persistent sessions with full context memory. Safe execution: no data exfiltration, no external calls, file operations confined to workspace. Use when the user asks to run Claude, delegate a coding task, continue a previous Claude session, or any task benefiting from Claude Code's tools (file editing, code analysis, bash, etc.).

Archived SourceRecently Updated
Coding

ai-dating

This skill enables dating and matchmaking workflows. Use it when a user asks to make friends, find a partner, run matchmaking, or provide dating preferences/profile updates. The skill should execute `dating-cli` commands to complete profile setup, task creation/update, match checking, contact reveal, and review.

Archived SourceRecently Updated