Agent Skillsdartsim/dart › dart-resume

dart-resume

GitHub

用于恢复和继续未完成开发任务的技能。它作为完成导向的任务管理器,负责重建执行计划、拆分任务包、验证结果并持续工作直至任务闭环或遇到阻塞。

.agents/skills/dart-resume/SKILL.md dartsim/dart

Trigger Scenarios

用户要求继续之前的中断工作 需要恢复未完成的开发任务

Install

npx skills add dartsim/dart --skill dart-resume -g -y
More Options

Non-standard path

npx skills add https://github.com/dartsim/dart/tree/main/.agents/skills/dart-resume -g -y

Use without installing

npx skills use dartsim/dart@dart-resume

指定 Agent (Claude Code)

npx skills add dartsim/dart --skill dart-resume -a claude-code -g -y

安装 repo 全部 skill

npx skills add dartsim/dart --all -g -y

预览 repo 内 skill

npx skills add dartsim/dart --list

SKILL.md

Frontmatter
{
    "name": "dart-resume",
    "description": "DART Resume: continue work from a previous session"
}

dart-resume

Use this skill in Codex to run the DART dart-resume workflow. The editable workflow source lives in .claude/commands/; this file is its generated adapter in the shared .agents/skills/ catalog.

Invocation

  • Claude Code: /dart-resume <arguments>
  • Codex: $dart-resume <arguments>

Treat the text after the skill name as $ARGUMENTS. When the workflow references $1, $2, etc., map those to the positional values supplied by the user.

Command Body

Resume unfinished work: $ARGUMENTS

Objective

dart-resume is a completion-oriented task manager, not a status lookup or a single-slice helper by default. Resume the named or reconstructed work, build or refresh the execution plan, track progress in the repo-owned task surface, split independent work into verifiable packets, use subagents/sub-sessions only when the user explicitly requested delegation and the current AI surface permits it, verify every result, and keep going until the whole task is complete or a real blocker or approval boundary remains.

Decisions must be evidence-based. Before choosing between meaningful options, first improve or define the verification/debugging method so it can catch false positives and false negatives. Use repository inspection, focused tests, benchmarks, A/B comparisons, GUI or visual evidence, logs, and external resource searches as needed to decide from evidence instead of preference.

For a docs/dev_tasks/<task> target, full completion means all feasible task work is finished, durable decisions and deferred work are promoted, and the temporary dev-task folder is removed in the completing change. Do not stop after one successful slice unless the user explicitly requested a limited mode.

Argument Handling

Use $ARGUMENTS to identify the target, explicit scope limits, and execution modifiers. Interpret arguments in this order:

  1. Explicit scope-limited mode: if arguments include status, audit-only, plan-only, slice, or next-slice, honor that limit and state it in the plan. Without one of these limiters, do not downscope the session to a single slice.
  2. Execution modifiers: if arguments include no-subagents, run serially but keep the same whole-task completion objective. If arguments explicitly request subagents, parallel, delegation, or workers, use available AI-native delegation when the current surface permits it. Tool-use modifiers do not limit scope unless paired with an explicit scope-limited mode.
  3. Explicit target path: if an argument names docs/dev_tasks/<task> or a file under that folder, treat that dev task as the target and read its current README.md and RESUME.md snapshots after recon, following the selective-reading protocol in docs/dev_tasks/README.md. Resolve relative paths from the repo root discovered by git rev-parse --show-toplevel, not from the launch directory.
  4. Closeout wording: words such as complete, finish, retire, close out, or cleanup are accepted but not required for dev-task completion. Treat retire / close out as emphasis that the likely end state is durable-doc promotion plus folder removal.
  5. Branch/PR/issue hint: otherwise, treat arguments as a branch, PR, issue, topic, or free-form resume hint and reconstruct the task from live repo state.

Examples (replace <task> with the selected task directory):

$dart-resume docs/dev_tasks/<task>
$dart-resume docs/dev_tasks/<task> --slice
$dart-resume PR 2991

Required Reading

@AGENTS.md @docs/dev_tasks/README.md @docs/ai/verification.md

Load docs/onboarding/ci-cd.md when CI is part of the current action, docs/onboarding/contributing.md before branch/PR work, and docs/onboarding/changelog.md for the closeout changelog decision.

Workflow

Recon (no changes)

git rev-parse --show-toplevel
git status -sb && git branch -vv && git log -10 --oneline --decorate
git diff --stat && git stash list
gh pr list --head "$(git branch --show-current)"
gh pr status

Reconstruct

If $ARGUMENTS names a docs/dev_tasks/<task> path, use that folder as the target and inspect its current README.md and RESUME.md snapshots before choosing work. Follow current owner pointers; retrieve historical entries only for relevant uncertainty, per docs/dev_tasks/README.md. If the folder is already absent, verify whether it was retired on current main, identify the durable owner doc that replaced it, and report that no folder cleanup remains.

Otherwise infer the task from branch name, commits, diffs, issue/PR description, and any docs/dev_tasks/<task>/ state. If the goal is still unclear after recon, stop and ask.

Before editing, write explicit pass/fail criteria for the resumed task:

  • what exact dev-task folder, branch, PR, issue, or topic is being completed;
  • which remaining checklist items, decisions, or evidence must be resolved;
  • how progress will be tracked during this session and across any handoff;
  • which independent subtasks, if any, can be delegated when explicitly authorized or run as separate serial sessions without creating conflicting edits;
  • which verification/debugging method will make false positives and false negatives unlikely before implementation decisions are made;
  • which A/B tests, benchmarks, resource searches, GUI checks, or textual checks are needed to support consequential decisions;
  • which durable docs will own any surviving decision or deferred work;
  • whether the resumed change may need a changelog entry and how the dart-changelog decision will be recorded;
  • which verification commands prove the result.

Continue

Continue the selected action within the reconstructed scope and supplied decisions. Use the task lifecycle in docs/dev_tasks/README.md, delegation contract in docs/ai/orchestration.md, and claim-specific gates/completion audit in docs/ai/verification.md. For 3D claims, use dart-verify-sim; retain the full solver/paper target across checkpoints.

Before closeout, run dart-changelog in decide or finalize mode. Promote durable dashboards/matrices and remaining-work decisions, then remove the completed task folder under its owner's retirement rules. If a real blocker remains, preserve exact current state and next action instead of claiming done.

For publication, load docs/onboarding/ai-reviews.md: merge the latest base before a PR push and reuse explicit authorization only within its recorded scope. Run pixi run lint before committing and the applicable pre-PR gates.

Safety

No destructive git commands (reset --hard, dropping stashes, deleting branches) without explicit maintainer/user approval.

Output

  • Reconstructed task and current branch/PR state
  • Plan followed and files changed
  • Verification commands run and their results
  • Completion-audit result: dev-task promotion/cleanup done or the remaining blocker
  • PR readiness, noting any external mutation that was explicitly approved

Version History

  • deb9869 Current 2026-09-08 21:46

    将调用命令从 OpenCode 更新为 Claude Code,并精简了 AI 测试范围。

  • bcd584e 2026-09-02 23:41
  • b9fbefc 2026-07-19 11:30

Same Skill Collection

.agents/skills/dart-analyze/SKILL.md
.agents/skills/dart-architecture/SKILL.md
.agents/skills/dart-audit-agent-compliance/SKILL.md
.agents/skills/dart-backport-pr/SKILL.md
.agents/skills/dart-benchmark-packet/SKILL.md
.agents/skills/dart-branch-cleanup/SKILL.md
.agents/skills/dart-build/SKILL.md
.agents/skills/dart-changelog/SKILL.md
.agents/skills/dart-ci/SKILL.md
.agents/skills/dart-close-issue/SKILL.md
.agents/skills/dart-contribute/SKILL.md
.agents/skills/dart-deps/SKILL.md
.agents/skills/dart-docs-update/SKILL.md
.agents/skills/dart-downstream-fix/SKILL.md
.agents/skills/dart-execute-packet/SKILL.md
.agents/skills/dart-fix-ci/SKILL.md
.agents/skills/dart-fix-issue/SKILL.md
.agents/skills/dart-io/SKILL.md
.agents/skills/dart-manage-pr/SKILL.md
.agents/skills/dart-mechanical-refactor/SKILL.md
.agents/skills/dart-model-upgrade/SKILL.md
.agents/skills/dart-new-task/SKILL.md
.agents/skills/dart-next/SKILL.md
.agents/skills/dart-plan-update/SKILL.md
.agents/skills/dart-pr/SKILL.md
.agents/skills/dart-python/SKILL.md
.agents/skills/dart-references/SKILL.md
.agents/skills/dart-release-merge-main/SKILL.md
.agents/skills/dart-release-packaging/SKILL.md
.agents/skills/dart-retro/SKILL.md
.agents/skills/dart-review-pr/SKILL.md
.agents/skills/dart-test/SKILL.md
.agents/skills/dart-triage-issue/SKILL.md
.agents/skills/dart-ultrawork/SKILL.md
.agents/skills/dart-verify-sim/SKILL.md
.claude/skills/dart-architecture/SKILL.md
.claude/skills/dart-build/SKILL.md
.claude/skills/dart-ci/SKILL.md
.claude/skills/dart-contribute/SKILL.md
.claude/skills/dart-io/SKILL.md
.claude/skills/dart-python/SKILL.md
.claude/skills/dart-references/SKILL.md
.claude/skills/dart-test/SKILL.md
.claude/skills/dart-verify-sim/SKILL.md
.codex/skills/dart-analyze/SKILL.md
.codex/skills/dart-architecture/SKILL.md
.codex/skills/dart-audit-agent-compliance/SKILL.md
.codex/skills/dart-backport-pr/SKILL.md
.codex/skills/dart-benchmark-packet/SKILL.md

Metadata

Files
0
Version
deb9869
Hash
74d918b6
Indexed
2026-07-19 11:30

Главная - Вики-сайт
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-09-10 13:01
浙ICP备14020137号-1 $Гость$