create-task
Skill4 starsChanged 20 days agoMITView on GitHub
Create a ClickUp task from raw input by classifying it as US, BUG, IMP, TASK, or SPIKE and filling the template.
13 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- fitlab-ai/agent-infratheir own version, first seen Oct 27, 2026
- luckyfoxdesign/llm-settings-templatetheir own version, first seen Oct 15, 2026
- cerbos/skillstheir own version, first seen Oct 15, 2026
- Jaxsonman/Crucibletheir own version, first seen Oct 13, 2026
- thiennq/skillstheir own version, first seen Oct 13, 2026
- ModelEngine-Group/fit-frameworktheir own version, first seen Oct 8, 2026
Used with
Projects with it also use harbor-exec and rewardkit.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Arize-aiSkillharbor-exec
Use when working with Harbor's `harbor exec` CLI workflow: compiling files, directories, or globs into Harbor tasks; running map jobs; configuring artifacts and existence-only verification; using map-reduce; writing or reviewing ExecConfig YAML/JSON/TOML; or debugging command behavior, config validation, and job outputs.
Also in 4 projects that use this
Arize-aiSkillrewardkit
Write Harbor task verifiers using Reward Kit. Use when creating or editing a
Also in 4 projects that use this
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- zahardev/aicontext★ 210% same
Wrote their own version · 12
- Arize-ai/phoenix★ 12ksame name
- harbor-framework/harbor★ 6.9ksame name
- laude-institute/harbor★ 6.9ksame name
- ModelEngine-Group/fit-framework★ 2.1ksame name
- fitlab-ai/agent-infra★ 87same name
- cerbos/skills★ 5same name
- dazzlerkumar/SKILLS★ 1same name
- gioe/tusk★ 1same name
- nordinit/agent-hq★ 1same name
- thiennq/skills★ 1same name
2 more not listed
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 repository45 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- a11y-auditorSkill
- api-contract-reviewSkill
- backend-releaseSkill
- bug-diagnosisSkill
- client-copy-reviewSkill
- codebase-designSkill
- code-reviewSkill
- create-issueSkill
- create-prSkill
- dependency-upgradeSkill
- deploy-evidenceSkill
- design-expertSkill
- design-system-docsSkill
- design-system-setupSkill
- domain-modelingSkill
- e2e-verifySkill
- evidence-auditSkill
- feature-discoverySkill
- frontend-releaseSkill
- generate-readmeSkill
- handoffSkill
- implement-taskSkill
- incident-responseSkill
- init-projectSkill
- live-docsSkill
- mcp-debuggingSkill
- memory-curatorSkill
- plan-expertSkill
- planning-featuresSkill
- project-memory-initSkill
- project-memory-refreshSkill
- project-profileSkill
- project-releaseSkill
- proof-runnerSkill
- release-checklistSkill
- researchSkill
- security-hardeningSkill
- security-reviewSkill
- tddSkill
- wiki-forgeSkill
- wiki-initSkill
- wiki-querySkill
- wiki-syncSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
16 rules. 0 are common, 10 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
10
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 19
Never invent information not present in the input — mark missing critical fields with ⚠ MISSING: [what's missing].
- Distinctive: line 20
Do not include technical implementation details in functional tasks (US, IMP) — that belongs in Technical Notes or separate subtasks.
- Distinctive: line 22
If the input is ambiguous in scope, take the most conservative interpretation and add an "Open Questions" section at the end.
- Distinctive: line 23
Use direct language, no filler.
- Distinctive: line 24
Every ticket must end with a "Summary" section: max 3 lines covering what to do, why, and how to verify it's done.
- Very specific: line 36
If no input is provided: use AskUserQuestion with:
- Very specific: line 44
If --type was explicitly provided, use that type and skip classification.
- Distinctive: line 56
If the input clearly maps to one type, proceed silently.
- Distinctive: line 58
If the input is ambiguous between two types (e.g., IMP vs US), state your classification and reasoning in one line before continuing:
- Very specific: line 75
Use the Read tool to load the template.
- Very specific: line 83
Required fields: must be filled. If the input does not provide enough information, mark the field with ⚠ MISSING: [description of what's …
- Distinctive: line 85
AC criteria: write each criterion as a functional, testable statement a QA engineer can verify independently.
- Distinctive: line 86
"Summary": always include this section at the end.
- Distinctive: line 87
Open questions: if the input is ambiguous about scope, add an "Open Questions" section at the end listing each uncertainty as a numbered …
- Distinctive: line 121
Create the task using:
- Very specific: line 149
If any required fields were marked with ⚠ MISSING:, include a reminder:
“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
create-task
Role: Senior Issue Triager following the agency's unified task creation standard.
Goal: Transform any raw input — a client message, a rough idea, a transcribed audio — into a well-structured ClickUp task using the correct template and the master prompt rules from the agency's task creation guide (Version 1.0, June 2026).
Master Prompt Rules (always apply)
- Title always starts with the task type in brackets:
[US],[BUG],[IMP],[TASK], or[SPIKE]. - Never invent information not present in the input — mark missing critical fields with
⚠ MISSING: [what's missing]. - Do not include technical implementation details in functional tasks (US, IMP) — that belongs in Technical Notes or separate subtasks.
- AC criteria must be functional and testable. Avoid vague criteria like "works correctly".
- If the input is ambiguous in scope, take the most conservative interpretation and add an "Open Questions" section at the end.
- Use direct language, no filler. The ticket must be taken by a dev without prior meetings.
- Every ticket must end with a "Summary" section: max 3 lines covering what to do, why, and how to verify it's done.
Step 1 — Gather Input
Parse $ARGUMENTS for:
--input "<text>"— the raw description to process--type <US|BUG|IMP|TASK|SPIKE>— force a specific task type (skip classification)
If --input is provided: acknowledge briefly and proceed to Step 2.
If no input is provided: use AskUserQuestion with:
- Header: "Create Task"
- Question: "Describe the task, bug, or idea you want to capture. You can paste a client message, a rough note, or a voice transcription — I'll handle the formatting."
Step 2 — Classify the Task Type
If --type was explicitly provided, use that type and skip classification.
Otherwise, analyze the raw input and classify it into one of the 5 types using these rules:
| Type | When to use |
[US] | Describes a new feature or capability from a user's perspective. Focus is on what the user wants to achieve. |
[BUG] | Describes something that is broken, behaves unexpectedly, or produces an error. |
[IMP] | Describes an improvement, change, or redesign of an existing feature. Not a new feature and not a bug. |
[TASK] | Describes internal technical work: refactoring, infrastructure, tooling, migrations, configuration. No user-facing story. |
[SPIKE] | Describes an investigation or research effort needed before committing to implementation. Centers on a specific question. |
If the input clearly maps to one type, proceed silently.
If the input is ambiguous between two types (e.g., IMP vs US), state your classification and reasoning in one line before continuing:
"I'm classifying this as
[IMP]because it modifies an existing feature rather than introducing new functionality."
Step 3 — Read the Corresponding Template
Based on the classified type, read the template file:
| Type | Template file |
[US] | templates/clickup/us_task_template.md |
[BUG] | templates/clickup/bug_task_template.md |
[IMP] | templates/clickup/imp_task_template.md |
[TASK] | templates/clickup/task_task_template.md |
[SPIKE] | templates/clickup/spike_task_template.md |
Use the Read tool to load the template. This is the exact structure that must be used — do not invent or remove sections.
Step 4 — Fill Out the Template
Populate every field in the template using only information present in the input. Apply the master prompt rules strictly:
- Required fields: must be filled. If the input does not provide enough information, mark the field with
⚠ MISSING: [description of what's missing]— never leave a required field blank or invent content. - Optional fields: fill them only if the input contains relevant information. If there is nothing to say, omit the entire optional section from the output.
- AC criteria: write each criterion as a functional, testable statement a QA engineer can verify independently.
- "Summary": always include this section at the end. Summarize in 3 lines max: what to do, why it matters, and how to confirm it's done.
- Open questions: if the input is ambiguous about scope, add an "Open Questions" section at the end listing each uncertainty as a numbered question.
Produce the fully filled-out ticket as a markdown block.
Step 5 — Confirm with User
Present the filled-out ticket to the user and ask:
"Here is the ticket I'm going to create in ClickUp. Does this look correct? Let me know if you'd like any changes before I create it."
Incorporate any requested changes. Repeat if necessary until the user confirms.
Step 6 — Select Target List in ClickUp
Fetch the workspace hierarchy to identify the correct list:
mcp__clickup__clickup_get_workspace_hierarchyPresent the available spaces and lists in a readable format. Use AskUserQuestion to ask:
"Which ClickUp list should I create this ticket in?"
Skip this step if the user already specified a list name or ID in the original input or arguments.
Step 7 — Create the Task in ClickUp
Create the task using:
mcp__clickup__clickup_create_task {
list_id: "<selected list id>",
name: "<ticket title>",
description: "<full ticket body in markdown — all sections>"
}The description must contain the complete filled-out template — all sections in order, exactly as confirmed in Step 5. Do not truncate, summarize, or omit any section.
Extract the url from the response.
Step 8 — Report
Output the result in this exact format:
✅ Task successfully created in ClickUp.
Type: <[US] | [BUG] | [IMP] | [TASK] | [SPIKE]>
Title: <task title>
URL: <clickup task url>If any required fields were marked with ⚠ MISSING:, include a reminder:
⚠ Incomplete fields: some required fields could not be filled with the available information.
Review the ticket and complete the fields marked with ⚠ MISSING before assigning it.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
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.

