auto
Run a task autonomously with reversible defaults and a reviewable handback.
Skill by KodingDev
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/kodingdev-auto
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/auto
What's inside
- Skill name
auto- Description
- Run a task autonomously with reversible defaults and a reviewable handback.
- License
- MIT
- Instructions
- 30 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
# Auto `/auto` is an explicit user-owned modifier. Everything after the invocation is the task. If no task is supplied, ask what to run and stop. ## Authority Invocation authorizes Meridian to: - choose reasonable reversible implementation details; - continue through ordinary friction without waiting for approval; - use appropriate internal disciplines and delegation; - produce the artifacts required by the task; and - commit completed in-scope code before handback. It does not authorize scope expansion, push, deploy, force-push, destructive data operations, credential or billing changes, shared-branch rewriting, or merge. ## Operating discipline - Ask only when the missing answer is human-only, materially changes scope or architecture, or gates an irreversible action. - Finish the requested outcome rather than starting adjacent work. - Keep legacy behavior intact when a migration cannot be completed safely. - Treat new user messages as current authority: absorb constraints, stop withdrawn scope, and end autonomous operation when the user takes back control. - Commit only files belonging to the task, per `meridian:commit`. - Stop when the professional standard's completion condition is met, not when elapsed time feels sufficient and not after inventing extra work. ## Handback Report what shipped, verification evidence, commits created, defaults chosen, unresolved decisions, and genuine blockers. Keep the handback concise and reviewable.
From KodingDev/meridian, MIT.
More from KodingDev/meridian
- 0.534score
triangulate
KodingDev
Use when credible sources conflict, one artifact cannot establish a load-bearing claim, a claim gates a hard-to-reverse action, or the user challenges it.
SkillCFreshness: C★ 5Claude Code +9 - 0.529score
commit
KodingDev
Commit message and pull request description conventions, and the clean-commit flow.
SkillCFreshness: C★ 5Claude Code +9 - 0.528score
sketch
KodingDev
Create and approve a lightweight implementation sketch for a small change.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
brainstorm
KodingDev
Turn an idea into an approved implementation spec through collaborative design.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
document
KodingDev
Capture hard-won engineering knowledge in durable project documentation.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
execute
KodingDev
Implement an approved spec or clearly defined piece of work.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
delegate
KodingDev
Isolate independent or context-heavy work when parallelism or fresh context materially improves the task.
SkillCFreshness: C★ 5Claude Code +9 - 0.520score
review
KodingDev
Run an independent, report-only review of a change.
SkillCFreshness: C★ 5Claude Code +9 - 0.518score
research
KodingDev
Verify unstable or unfamiliar external API and library contracts that materially affect the task.
SkillCFreshness: C★ 5Claude Code +9 - 0.518score
respond
KodingDev
Evaluate and address concrete code-review feedback against repository reality.
SkillCFreshness: C★ 5Claude Code +9 - 0.518score
debug
KodingDev
Diagnose bugs, test failures, and unexpected behavior from evidence before changing code.
SkillCFreshness: C★ 5Claude Code +9
Signals
- Stars
- 5
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 25, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/KodingDev/meridian
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from KodingDev
- 0.534score
triangulate
KodingDev
Use when credible sources conflict, one artifact cannot establish a load-bearing claim, a claim gates a hard-to-reverse action, or the user challenges it.
SkillCFreshness: C★ 5Claude Code +9 - 0.529score
commit
KodingDev
Commit message and pull request description conventions, and the clean-commit flow.
SkillCFreshness: C★ 5Claude Code +9 - 0.528score
sketch
KodingDev
Create and approve a lightweight implementation sketch for a small change.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
brainstorm
KodingDev
Turn an idea into an approved implementation spec through collaborative design.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
document
KodingDev
Capture hard-won engineering knowledge in durable project documentation.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
execute
KodingDev
Implement an approved spec or clearly defined piece of work.
SkillCFreshness: C★ 5Claude Code +9
Cite this page
auto skill. Agent Mart. https://agentmart.dev/skills/kodingdev-auto (last updated 2026-10-24)