agentleFS
zeng-xiangwangzeng-xiangwang/trae-skills/manage-taskboard/SKILL.md

manage-taskboard

Skill0 starsChanged 3 months agoMITView on GitHub

Manage taskboard projects, issues, issue relations, and comments through the taskctl CLI. Use when the agent needs to track a new requirement, inspect project work, create or update issues, relate dependent work, add progress notes, begin work on an issue, record completion, or coordinate concurrent updates.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. chuspeeism/dashi-taskboard★ 3.5ksame name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 7 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use taskctl for every project, issue, and comment operation.

    line 8
  2. Very specific:

    Search for an existing issue before creating one.

    line 18
  3. Distinctive:

    If an issue already tracks the same requirement, append the new requirement or acceptance detail to that issue without discarding its …

    line 19
  4. Distinctive:

    If the work depends on, blocks, is blocked by, or is closely related to another issue, add the matching issue relation.

    line 20
  5. Distinctive:

    Use a parent/sub-issue relation when one requirement is a contained part of a larger issue.

    line 21
  6. Distinctive:

    Create a new issue only when no existing issue reasonably tracks the requirement.

    line 22
  7. Distinctive:

    Do not create, append, or relate a tiny or trivial request that does not benefit from durable tracking.

    line 23
  8. Distinctive:

    Before executing an issue, read the latest issue content and all comments.

    line 24
  9. Very specific:

    When understanding that image is necessary, use attachment download to save it locally, then inspect the saved file with an available …

    line 26
  10. Very specific:

    Create or update issues with the CLI.

    line 27
  11. Distinctive:

    Before requesting review, verify the requested work and acceptance criteria.

    line 32
  12. Very specific:

    After implementation and self-verification, add a comment summarizing the key changes, verification, result, and remaining risks.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Manage Taskboard

Use taskctl for every project, issue, and comment operation. Read references/cli.md before choosing a command or option.

Prerequisites

The local Taskboard service must be running. Default address is http://127.0.0.1:47823, override with the environment variable CODEX_TASKBOARD_URL. Start it from the Taskboard repo root with npm start (run npm install first if dependencies are missing). The SQLite database lives at .data/taskboard.sqlite. If taskctl reports the service unavailable (exit code 3), start the service before retrying.

Before first use: set <TASKCTL_CLI_PATH> in your project rules (absolute path to taskctl.mjs in the Taskboard repo). If you've linked it globally you can just call taskctl directly.

Workflow

  1. Search for an existing issue before creating one. Use context current, then list the project issues and compare their identifiers, titles, descriptions, and status.
    • If an issue already tracks the same requirement, append the new requirement or acceptance detail to that issue without discarding its existing scope.
    • If the work depends on, blocks, is blocked by, or is closely related to another issue, add the matching issue relation.
    • Use a parent/sub-issue relation when one requirement is a contained part of a larger issue. A child has one parent; a parent may have many sub-issues.
    • Create a new issue only when no existing issue reasonably tracks the requirement.
    • Do not create, append, or relate a tiny or trivial request that does not benefit from durable tracking.
  2. Before executing an issue, read the latest issue content and all comments. Treat comments as part of the current requirements, especially when completed work has been returned for changes.
    • In a description or comment, ![alt](/api/attachments/<id>/content) marks an inline image at that exact position in the text.
    • When understanding that image is necessary, use attachment download to save it locally, then inspect the saved file with an available image-viewing tool.
  3. Create or update issues with the CLI; consume its JSON output. Issues created through taskctl are assigned to Codex Agent by default (a server-side actor record). Later CLI updates do not change the assignee.
  4. Attribute every issue, relation, or comment mutation to the current Trae Work conversation by passing --thread-id ID explicitly. Trae Work does not set CODEX_THREAD_ID, so the explicit flag is required on every write command. Read commands do not require a thread id.
  5. To claim a todo issue, move it to in_progress with --if-version from the latest read before starting implementation. If this claim reports a version conflict or a new read shows that its status changed, skip the issue and do not implement it.
  6. Include --if-version <version> on every concurrent update, using the version returned by the latest read.
  7. Before requesting review, verify the requested work and acceptance criteria.
  8. After implementation and self-verification, add a comment summarizing the key changes, verification, result, and remaining risks; then move the issue to in_review. Never move it directly to done.
  9. Move an issue from in_review to done only when the user explicitly confirms acceptance or explicitly asks to mark it complete. Self-verification alone is not sufficient.
  10. Move work that cannot continue to blocked, and work that will not continue to canceled.

For version conflicts outside the initial claim, read the issue again, reconcile the newer state, and retry with its current version.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.