Agent skill

Kanban Spec

by cyanluna-git in cyanluna-git/cyanluna.skills

Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task.

MITAuto-check passedProductivity & Automation

Install Kanban Spec

skills CLI
$ npx skills add cyanluna-git/cyanluna.skills --skill kanban-spec -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install cyanluna-git/cyanluna.skills kanban-spec --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/cyanluna-git/cyanluna.skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/kanban-spec .claude/skills/kanban-spec && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
kanban-spec
GitHub stars
183
Token cost
~1.4k tokens
SKILL.md length
532 words
Files
1
Skills in repo
16
Repo updated
First seen
Licence
MIT

At a glance

Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task.

  • Works in 5 steps: Problem Framing → Scope Definition → Success Metrics → …
  • Starting something new that needs proper scoping before refinement
  • Calls python3
  • Tasks that involve Task management

What it does

Kanban Spec is an agent skill from cyanluna-git/cyanluna.skills. Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task. Use when starting something new that needs proper scoping before refinement. Complements /kanban-refine (which refines existing tasks).

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Productivity & Automation, covering Task management. The licence is MIT.

When your agent uses it

  • Starting something new that needs proper scoping before refinement
  • Tasks that involve Task management

Example prompts

  • “/kanban-spec”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Problem Framing
  2. Scope Definition
  3. Success Metrics
  4. Edge Cases & Risks
  5. Acceptance Criteria

What it can do on your machine

Read from SKILL.md and the folder at commit df5be37. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Kanban Spec loads about 1.4k tokens when it runs. Until then it costs about 66 tokens; SKILL.md has 532 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~66
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from cyanluna-git/cyanluna.skills at commit df5be37, republished under its MIT licence (© cyanluna-git). 532 words, ~1,390 tokens.

Download SKILL.mdSave it as .claude/skills/kanban-spec/SKILL.md (or your agent's skills folder).
name
kanban-spec
description
Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task. Use when starting something new that needs proper scoping before refinement. Complements /kanban-refine (which refines existing tasks).
license
MIT

Shared context: read ../kanban/shared.md for DB path and task creation format. Safety principles: read ../kanban/principles.md — mandatory, not optional.

/kanban-spec [title] — New Feature Spec

Turns a rough idea into a structured spec through 5 phases, then optionally creates a kanban task.

When to use this vs /kanban-refine:

  • /kanban-spec — Starting from scratch. No task exists yet. Idea is vague.
  • /kanban-refine <ID> — Task already exists in backlog. Needs requirement clarification.

Phase 1 — Problem Framing

Goal: Understand the real problem before discussing solutions.

Ask the user:

  1. What user pain or business need is this solving?
  2. Who experiences this problem? (user type, frequency)
  3. What happens today without this feature? (workaround or nothing?)
  4. Why now — what's prompting this? (deadline, user feedback, blocker?)

Extract and record:

PROBLEM_STATEMENT = one sentence: "[User type] can't [do X] because [root cause], causing [impact]"

If the user can't articulate a clear problem, probe further before proceeding. Do not move to Phase 2 until the problem is clear.


Phase 2 — Scope Definition

Goal: Draw a hard boundary around what's in and out.

Ask:

  1. What is the minimal version of this that solves the problem? (MVP scope)
  2. What would be nice but is NOT required for this task?
  3. Does this replace or extend existing functionality?
  4. Any hard constraints? (deadline, tech stack, team size, API limits)

Record:

SCOPE_IN  = bulleted list of what's included
SCOPE_OUT = bulleted list of explicit exclusions
CONSTRAINTS = technical, time, or team constraints

Call out scope creep early: if the user mentions a "nice to have" that doubles the work, flag it and suggest splitting into a separate task.


Phase 3 — Success Metrics

Goal: Define what "done and working" looks like from the user's perspective.

Ask:

  1. How will you know this is working? (observable outcome)
  2. Is there a measurable target? (latency, conversion rate, error rate, etc.)
  3. What's the minimum bar for the first version? What would make it great?

Record:

METRICS = list of observable, preferably measurable outcomes

At least one metric must be verifiable — "it works" is not a metric.


Show full SKILL.md (232 more words)Show less
Phase 4 — Edge Cases & Risks

Goal: Surface the 20% of cases that cause 80% of bugs.

Based on the scope defined, identify:

  • Empty states: What if there's no data? First-time user?
  • Error states: What if an external dependency fails?
  • Boundary conditions: Limits, maximums, empty inputs
  • Concurrent access: Multiple users, race conditions
  • Rollback: Can this be undone? What's the data impact?
  • Security: Does this handle untrusted input? Auth required?

Ask the user to confirm edge cases are understood, or add any you've missed.

Record: EDGE_CASES = list of identified edge cases


Phase 5 — Acceptance Criteria

Goal: Write the checklist that Builder and Shield will use to verify completion.

Synthesize everything from phases 1–4 into verifiable criteria:

## Acceptance Criteria
- [ ] [Functional: behavior visible to user]
- [ ] [Error handling: specific edge case is handled]
- [ ] [Performance: if metric defined]
- [ ] [Security: if auth/input involved]
- [ ] [Regression: existing behavior preserved]

Rules:

  • Each criterion must be independently verifiable
  • Use active voice: "User can X" not "X is implemented"
  • No vague items like "works correctly" or "handles errors"
  • 3–8 criteria is the right range; more than 8 means scope is too broad

Output: Full Spec

Present the complete spec to the user:

markdown
## Spec: [title]

### Problem
[PROBLEM_STATEMENT]

### Scope
**In:**
[SCOPE_IN]

**Out:**
[SCOPE_OUT]

**Constraints:** [CONSTRAINTS or "None identified"]

### Success Metrics
[METRICS]

### Edge Cases
[EDGE_CASES]

### Acceptance Criteria
[ACCEPTANCE_CRITERIA]

Ask the user: [c] Create kanban task / [e] Edit spec / [x] Discard


Create Kanban Task (if user selects [c])

Ask two final questions via AskUserQuestion:

  1. Priority: critical / high / medium / low
  2. Level: L1 (quick, <1h) / L2 (standard, <1d) / L3 (complex, multi-day)

Then read DB path from shared.md and create the task:

bash
CONFIG=$(cat .claude/kanban.json 2>/dev/null || cat .codex/kanban.json 2>/dev/null)
PROJECT=$(echo "$CONFIG" | python3 -c "import sys,json; d=json.load(sys.stdin); print(d['project'])" 2>/dev/null || basename "$(pwd)")
DB="$HOME/.claude/kanban-dbs/${PROJECT}.db"

DESCRIPTION="## Goal\n[PROBLEM_STATEMENT]\n\n## Scope\n**In:**\n[SCOPE_IN]\n\n**Out:**\n[SCOPE_OUT]\n\n## Acceptance Criteria\n[ACCEPTANCE_CRITERIA]\n\n## Edge Cases\n[EDGE_CASES]"

python3 - <<PY
import sqlite3 as sq
conn = sq.connect("$DB")
cur = conn.execute(
    "INSERT INTO tasks (project, title, description, priority, level, status, tags) VALUES (?, ?, ?, ?, ?, 'todo', '[]')",
    ("$PROJECT", "[title]", """$DESCRIPTION""", "[priority]", [level_number])
)
task_id = cur.lastrowid
conn.commit()
conn.close()
print(task_id)
PY

Output: "✅ Task #[ID] created: [title] → Run /kanban-refine [ID] to sharpen details, or /kanban-run [ID] to implement."

© cyanluna-git, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in kanban-spec of cyanluna-git/cyanluna.skills.

Open the folder on GitHubat commit df5be37

Compare with similar skills

Kanban Spec next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Kanban Spec compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Kanban Spec this skillcyanluna-git/cyanluna.skills183—~1.4kAutomated safety check: PassMIT
Superset Agent Standupsuperset-sh/superset15k—~712Automated safety check: PassCustom licence
AgentRQ Workspace Agentagentrq/agentrq1.1k—~1.9kAutomated safety check: PassAGPL-3.0
Markdown Task Managerioniks/MarkdownTaskManager535—~2.2kAutomated safety check: PassMPL-2.0
Pi Messenger Crewnicobailon/pi-messenger720—~3.7kAutomated safety check: PassNone
Codekanban CLIfy0/CodeKanban225—~2.7kAutomated safety check: PassApache-2.0

Similar skills

  • Superset Agent Standup

    superset-sh/superset

    Sweeps every Superset workspace, task and agent terminal to report what finished, what needs review and what is blocked, read-only, and can publish the digest as a page.

    15k GitHub stars~712 tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Guides a workspace agent through executing assigned tasks, replying to a remote human operator, and creating sub-tasks, memory and events inside an AgentRQ workspace.

    1.1k GitHub stars~1.9k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Markdown Task Manager

    ioniks/MarkdownTaskManager

    A skill your agent uses when managing tasks, the system is a Kanban task manager based on local Markdown files (kanban.md and archive.md).

    535 GitHub stars~2.2k tokensUpdated 3 mo ago
    Productivity & AutomationAuto-check passed
  • Pi Messenger Crew

    nicobailon/pi-messenger

    Orchestrator reference for pi-messenger Crew planning, task management, configuration, and agent coordination.

    720 GitHub stars~3.7k tokensUpdated 1 mo ago
    Productivity & AutomationAuto-check passed
  • Codekanban CLI

    fy0/CodeKanban

    Operate CodeKanban workflows, terminal sessions, and web sessions through the installable codekanban-cli command.

    225 GitHub stars~2.7k tokensUpdated 6 days ago
    Productivity & AutomationAuto-check passed
  • Kanban Video Orchestrator

    Luciole-Studio/Misaka-Agent

    Plan and run multi-agent video production pipelines. An agent skill from Luciole-Studio/Misaka-Agent.

    158 GitHub starsUsed in 2 repos~2.4k tokens
    Productivity & AutomationAuto-check: notes

More from cyanluna-git/cyanluna.skills

All 16 skills in this repo
  • Kanban Batch Run

    cyanluna-git/cyanluna.skills

    Run multiple kanban tasks end-to-end in Rolling Wave order — refine each task based on the prior card's actual implementation, then implement, then verify, then refine the next.

    183 GitHub stars~2.7k tokensUpdated 3 mo ago
    Auto-check passed
  • Kanban

    cyanluna-git/cyanluna.skills

    Manage project tasks in a local SQLite DB (~/.claude/kanban-dbs/{project}.db).

    183 GitHub stars~3.6k tokensUpdated 3 mo ago
    Auto-check passed
  • Kanban Explore

    cyanluna-git/cyanluna.skills

    Codebase exploration skill for uncertain implementation direction.

    183 GitHub stars~3k tokensUpdated 3 mo ago
    Auto-check passed
  • Kanban Local

    cyanluna-git/cyanluna.skills

    Local markdown-file kanban for toy and personal projects. An agent skill from cyanluna-git/cyanluna.skills.

    183 GitHub stars~1.9k tokensUpdated 3 mo ago
    Auto-check passed
  • Kanban Run

    cyanluna-git/cyanluna.skills

    Run the AI team pipeline for kanban tasks — orchestration loop with 6 agents (Planner, Critic, Builder, Shield, Inspector, Ranger), single-step execution, and code review.

    183 GitHub stars~4.3k tokensUpdated 3 mo ago
    Auto-check passed
  • Project Kickstart

    cyanluna-git/cyanluna.skills

    Full project pipeline — SRS → Plan → Tasks + TDD → Rolling Wave Execute.

    183 GitHub stars~1.8k tokensUpdated 3 mo ago
    Auto-check passed

Questions about Kanban Spec

What does Kanban Spec do?

Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task. skills. Turn vague intent into a precise, executable spec through 5 structured phases, then optionally create a kanban task.

When should I use Kanban Spec?

Kanban Spec fits situations like: starting something new that needs proper scoping before refinement; tasks that involve Task management.

How do I install Kanban Spec in Claude Code?

Run `npx skills add cyanluna-git/cyanluna.skills --skill kanban-spec -a claude-code`. Or copy the skill folder (kanban-spec in cyanluna-git/cyanluna.skills) into .claude/skills/kanban-spec in your project. Claude Code loads it when a task matches its description.

How do I install Kanban Spec in Codex?

Run `npx skills add cyanluna-git/cyanluna.skills --skill kanban-spec -a codex`. Or copy the skill folder (kanban-spec in cyanluna-git/cyanluna.skills) into .agents/skills/kanban-spec in your project. Codex loads it when a task matches its description.

Can I use Kanban Spec in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add cyanluna-git/cyanluna.skills --skill kanban-spec -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/kanban-spec, .gemini/skills/kanban-spec, .github/skills/kanban-spec and .opencode/skills/kanban-spec in your project.

What does Kanban Spec need to run?

Going by SKILL.md and its folder, Kanban Spec needs the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Kanban Spec access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Kanban Spec safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Kanban Spec use?

Kanban Spec is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Kanban Spec use?

About 1.4k tokens (SKILL.md is roughly 5.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Kanban Spec?

Skills that share tags, products or a category with Kanban Spec: Superset Agent Standup (superset-sh/superset, 15k stars), AgentRQ Workspace Agent (agentrq/agentrq, 1.1k stars), Markdown Task Manager (ioniks/MarkdownTaskManager, 535 stars) and Pi Messenger Crew (nicobailon/pi-messenger, 720 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Kanban Spec?

cyanluna-git (a GitHub user) maintains it in cyanluna-git/cyanluna.skills, which has 183 GitHub stars. The repository holds 16 skills in this directory. The repository was last updated on June 22, 2026.

Source: cyanluna-git/cyanluna.skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.