Agent skill

Create Issue

by foyzulkarim in foyzulkarim/claude-lens

File GitHub issues for the claude-lens project with the right shape for the work type — plan tasks from specs/claude-lens-plan.md, Phase 4 page tasks, spikes, bugs, enhancements, chores.

MITAuto-check passedProduct & Project Management

Install Create Issue

skills CLI
$ npx skills add foyzulkarim/claude-lens --skill create-issue -a claude-code

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

GitHub CLI
$ gh skill install foyzulkarim/claude-lens create-issue --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/foyzulkarim/claude-lens.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/create-issue .claude/skills/create-issue && 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
create-issue
GitHub stars
250
Token cost
~2.4k tokens
SKILL.md length
1,016 words
Files
2 (incl. scripts)
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

File GitHub issues for the claude-lens project with the right shape for the work type — plan tasks from specs/claude-lens-plan.md, Phase 4 page tasks, spikes, bugs, enhancements, chores.

  • Works in 6 steps: Classify the request → Source the content → Title, labels, milestone → …
  • Asks to create/file/open an issue
  • SKILL.md covers Step 1 — Classify the request, Step 2 — Source the content, Step 3 — Title, labels,… and Step 4 — Body shapes, plus 2 more sections
  • Runs Shell scripts from its folder; calls gh and bash

What it does

Create Issue is an agent skill from foyzulkarim/claude-lens. File GitHub issues for the claude-lens project with the right shape for the work type — plan tasks from specs/claude-lens-plan.md, Phase 4 page tasks, spikes, bugs, enhancements, chores. Use this whenever the user asks to create/file/open an issue, turn a plan task (e.g. "P2-3") into an issue, scaffold issues for a phase, log a bug, or propose a spike/investigation — even if they don't say the word "issue" (e.g. "let's get P1 into GitHub", "track this as a task").

Its SKILL.md is about 2.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including scripts (for example `scripts/publish.sh`).

It sits in Product & Project Management. It works with GitHub. The repository describes itself as: A local dashboard for visualizing your Claude Code usage — sessions, token costs, cache performance, tool calls, and daily breakdowns. The licence is MIT.

When your agent uses it

  • Asks to create/file/open an issue
  • Turn a plan task (e.g

Example prompts

  • “t say the word”
  • “(e.g.”
  • “s get P1 into GitHub”
  • “/create-issue”

Requirements

  • A Bash shell

Workflow steps

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

  1. Classify the request
  2. Source the content
  3. Title, labels, milestone
  4. Body shapes
  5. Draft locally (never file directly)
  6. Publish in one run

What it can do on your machine

Read from SKILL.md and the folder at commit 7937ea1. 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

    Ships 1 file in scripts/ (Shell), which the agent can run.

    Shell commands in SKILL.md call:

    • gh
    • bash

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

  • Network

    No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.

    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

Create Issue loads about 2.4k tokens when it runs. Until then it costs about 121 tokens; SKILL.md has 1,016 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~121
When it runs · the whole SKILL.md, loaded when a task matches
~2.4k

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); the scripts in this folder are not scanned.

SKILL.md

The full file from foyzulkarim/claude-lens at commit 7937ea1, republished under its MIT licence (© foyzulkarim). 1,016 words, ~2,411 tokens.

Download SKILL.mdSave it as .claude/skills/create-issue/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
create-issue
description
File GitHub issues for the claude-lens project with the right shape for the work type — plan tasks from specs/claude-lens-plan.md, Phase 4 page tasks, spikes, bugs, enhancements, chores. Use this whenever the user asks to create/file/open an issue, turn a plan task (e.g. "#P2-3") into an issue, scaffold issues for a phase, log a bug, or propose a spike/investigation — even if they don't say the word "issue" (e.g. "let's get P1 into GitHub", "track this as a task").
model
inherit

Create Issue

Issues in this repo come in different shapes because the work does: a plan task carries acceptance criteria written months in advance; a bug carries a repro; a spike carries a question and a timebox. One rigid template can't serve all of these, so this skill generates the right body per type and files it with gh issue create.

The specs are the source of truth. Never invent scope or acceptance criteria — pull them from the docs, and if a plan task is missing something, flag the gap to the user instead of papering over it.

Step 1 — Classify the request

TypeSignals
plan-taskA task ID like #P2-3, or "file the Phase 1 issues" — anything that maps to a checkbox in specs/claude-lens-plan.md
pageA plan-task that is one of the 11 Phase 4 page tasks (#P4-2, 4–10, 14–16). Same as plan-task plus extra page-specific content (see Step 2)
spikeOpen question, investigation, "figure out whether…", timeboxed research. Not in the plan doc
bugSomething built is behaving wrongly
enhancementNew feature/improvement idea that is not in the plan doc. If it belongs in the plan, suggest adding it there first — the plan is the backlog of record
choreRepo/tooling/docs upkeep with no user-facing behavior

If the request maps to several plan tasks (e.g. a whole phase), file one issue per task, sequentially in plan order.

Step 2 — Source the content

plan-task / page: Read the task's entry in specs/claude-lens-plan.md — the task body is the scope, and the Acceptance: line must be copied verbatim into the issue (plan-doc rule). Follow the task's spec references before writing the body:

  • architecture §N → docs/claude-lens-architecture.md
  • pages §N → specs/claude-lens-pages.md
  • gates → specs/gates.md
  • mockups → specs/pages/<page>.html

Derive Depends on / Unblocks from plan order: phases 0–3 are strictly sequential; within Phase 4 the plan's ordering notes apply. Reference dependencies by task ID — don't hit GitHub during drafting; if a dependency was already filed, its issue number is in its own draft's frontmatter under specs/issues/.

page (extra): The issue must also carry:

  • The binding section list from the page's table in claude-lens-pages.md (spec wins over mockup).
  • The mockup path (specs/pages/X.html) as visual reference, not exhaustive contract.
  • Any of the six known spec-vs-mockup gaps that affect this page (listed in the plan's Phase 4 standing rules) — implement from the spec table.
  • The three standing rules as a Definition of Done: Cypress smoke spec (render + one drill-link), Storybook for component states, manual mockup sign-off on real data.

spike / bug / enhancement / chore: Content comes from the conversation. Before filing, make sure you have: for a bug, a repro and expected-vs-actual; for a spike, the question, why it matters now, and a timebox/exit criterion. If these are missing, ask — a bug without a repro or a spike without an exit criterion is not worth filing.

enhancement, when it's more than a one-liner: first check specs/requirements/ for a REQ doc covering it. If one exists, source the issue from it exactly as plan tasks are sourced from the plan doc — summary from its problem statement, acceptance criteria copied verbatim, REQ path linked under References. If none exists and the scope is fuzzy (multiple behaviors, unstated edge cases, no verifiable done-signal), pause and suggest the user run /plan-requirements first — it's their user-invocable interview skill (you cannot invoke it yourself) and produces the REQ doc this issue should be derived from. Only file straight from conversation when the enhancement is small enough that its acceptance line is obvious. This keeps one rule across all issue types: issues cite a requirements source; they never invent one.

Show full SKILL.md (427 more words)Show less

Step 3 — Title, labels, milestone

TypeTitleLabelsMilestone
plan-task / page#P<phase>-<n> — <task title from plan>phase-<N>Phase <N> — <name>
spikespike: <question in one line>spike + phase-<N> if it blocks a phasephase milestone if applicable
bugbug: <symptom>bug + phase-<N> of the affected areacurrent phase
enhancementfeat: <capability>enhancementusually none until planned
chorechore: <what>documentation/none as fitsusually none

Phase labels (phase-0…phase-5) and the six phase milestones already exist. The spike label may not — create it on first use: gh label create spike --description "Timeboxed investigation" --color 1d76db.

Step 4 — Body shapes

Shapes below; drop sections that would be empty rather than leaving placeholders. Reference dependencies by task ID — GitHub issue numbers usually don't exist yet at drafting time (a filed dependency's number is in its own draft's frontmatter if you need it).

plan-task / page

markdown
Task **#P<X>-<Y>** from [specs/claude-lens-plan.md](../blob/main/specs/claude-lens-plan.md) — Phase <X>.

## Summary
<one or two sentences: what this delivers and why it's in this phase>

## Scope
<bullets pulled from the task body; spec section refs inline, e.g. "per architecture §4">

## Acceptance criteria
<the plan doc's *Acceptance:* line, verbatim, as bullets>

## Dependencies
- Depends on: <task IDs / #issue numbers, or "none — first in phase">
- Unblocks: <next task(s)>

## References
<spec sections, mockup path, decisions-log rows that constrain this task>

For page tasks, add between Acceptance and Dependencies:

markdown
## Page contract (pages spec §<N>)
<the section table rows: section → data dep → tier behavior → drill target>
Spec-vs-mockup gaps to implement from the spec table: <the relevant items, or "none for this page">

## Definition of done (Phase 4 standing rules)
- [ ] Cypress smoke spec: route renders key sections from fixtures; one drill-link lands filtered
- [ ] Component states covered in Storybook (not Cypress)
- [ ] Manual visual sign-off vs `specs/pages/<page>.html` on real data; plan checkbox flipped

spike

markdown
## Question
<the single question this spike answers>

## Why now
<what decision or task is blocked on the answer>

## Approach
<how to investigate — prototype, measurement, reading; keep it cheap>

## Timebox & exit criterion
<e.g. "1 day; exits with a decisions-log row + recommendation, regardless of outcome">

bug

markdown
## Symptom
<what happens, where — page/route/module>

## Repro
<numbered steps, or the fixture/data that triggers it>

## Expected vs actual
<one line each>

## Suspected area
<file/module if known; omit if not>

enhancement / chore

markdown
## What & why
<the capability or upkeep, and the motivation>

## Acceptance
<how we'll know it's done — verifiable, in the spirit of the plan doc's acceptance lines;
copied verbatim from the REQ doc when one exists>

## References
<`specs/requirements/REQ-<slug>.md` if the issue was derived from one>

Step 5 — Draft locally (never file directly)

Issues are drafted as local files first so the user can edit them until happy, then published to GitHub in one batch — not one HTTP round-trip per revision. Write each draft to specs/issues/<ID>-<slug>.md (e.g. P0-2-move-v1-into-legacy.md; for ad-hoc types use bug-<slug>.md, spike-<slug>.md, …):

markdown
---
title: "#P0-2 — Move V1 app into legacy/"
labels: phase-0
milestone: Phase 0 — Spec closure & repo prep
status: draft
---

<issue body per Step 4>

Frontmatter rules: labels comma-separated; omit milestone if none; status is the lifecycle:

  • draft — being edited. The user owns drafts once written: before regenerating or updating one, re-read it — their manual edits must survive, so never overwrite except on their explicit ask.
  • ready — user has approved it for publishing. Flip drafts to ready only when the user says they're final ("publish these", "file them", "looks good, ship it").
  • filed — on GitHub; issue: and url: lines were added by the publish script. Filed drafts are frozen local records — GitHub is the source of truth from then on; don't edit them.

Step 6 — Publish in one run

When the user says the drafts are final, flip the approved ones to status: ready, then run the bundled script from the repo root:

bash
bash .claude/skills/create-issue/scripts/publish.sh

It files every ready draft sequentially in natural-sort order (P0-2 before P0-10) via gh issue create, marks each filed with its issue number and URL, and paces requests to respect GitHub rate limits. Afterwards, report the URLs.

  • Do not check off plan checkboxes on filing — boxes flip when issues close, per the plan doc's usage note. Flip [ ] to [~] only if the user says work is starting now.
  • If gh fails mid-run (auth, missing label/milestone), already-filed drafts are safely marked filed — fix the cause (e.g. gh label create spike …), rerun the script, and it resumes with the remaining ready drafts.

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

Files

SKILL.md and 1 other file (scripts) in .claude/skills/create-issue of foyzulkarim/claude-lens.

  • SKILL.md
  • scripts/publish.sh

Open the folder on GitHubat commit 7937ea1

Compare with similar skills

Create Issue 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.

Create Issue compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create Issue this skillfoyzulkarim/claude-lens250—~2.4kAutomated safety check: PassMIT
CCPM Project Managementautomazeio/ccpm8.4k—~1.1kAutomated safety check: PassMIT
Release ValidationMesh-LLM/mesh-llm3.5k—~2.6kAutomated safety check: PassApache-2.0
Final Release Reviewopenai/openai-agents-python30k—~5.4kAutomated safety check: PassMIT
Ouroboros PM InterviewQ00/ouroboros6.2k—~5.7kAutomated safety check: PassMIT
Project Managerpwrdrvr/openclaw-codex-app-server265—~1.5kAutomated safety check: PassMIT

Similar skills

  • Runs a spec-driven workflow from PRD to epic to GitHub issues to parallel agents, with status, standup and blocked-work reports from bundled scripts.

    8.4k GitHub stars~1.1k tokensUpdated 6 mo ago
    Product & Project ManagementAuto-check passed
  • Release Validation

    Mesh-LLM/mesh-llm

    A skill your agent uses when validating a MeshLLM release candidate or current HEAD against the last GitHub release, assembling the canonical feature/fix/modification inventory, testing locally…

    3.5k GitHub stars~2.6k tokensUpdated today
    Product & Project ManagementAuto-check passed
  • Final Release Review

    openai/openai-agents-python

    Official

    Assess a Python SDK release candidate or release plan against the previous release and recommend ship or block.

    30k GitHub stars~5.4k tokensUpdated today
    Product & Project ManagementAuto-check passed
  • Runs a guided product-manager interview that classifies each question automatically and produces a Product Requirements Document.

    6.2k GitHub stars~5.7k tokensUpdated yesterday
    Product & Project ManagementAuto-check passed
  • Project Manager

    pwrdrvr/openclaw-codex-app-server

    Manage GitHub issues and the GitHub Project board for the current repository, while keeping the local tracker in sync.

    265 GitHub stars~1.5k tokensUpdated 1 mo ago
    Product & Project ManagementAuto-check passed
  • Create Issue

    tradingstrategy-ai/frontend

    Creates a new GitHub issue with structured user story format, acceptance criteria, and project integration.

    150 GitHub stars~607 tokensUpdated yesterday
    Product & Project ManagementAuto-check passed

More from foyzulkarim/claude-lens

  • Archive Issue

    foyzulkarim/claude-lens

    Retire a closed issue's working artifacts out of specs/ into the GitHub wiki — use when the user asks to archive a finished issue, empty out specs/ for a done task, or move an issue's…

    250 GitHub stars~2.8k tokensUpdated 2 mo ago
    Auto-check passed
  • Move To Worktree

    foyzulkarim/claude-lens

    After /start-task: park the current clean, pushed feature branch in its own issue-numbered nested worktree (.worktrees/<issue) and return the primary checkout to current main, so the next parallel…

    250 GitHub stars~663 tokensUpdated 2 mo ago
    Auto-check: notes

Works with

Questions about Create Issue

What does Create Issue do?

File GitHub issues for the claude-lens project with the right shape for the work type — plan tasks from specs/claude-lens-plan.md, Phase 4 page tasks, spikes, bugs, enhancements, chores. Create Issue is an agent skill from foyzulkarim/claude-lens.md, Phase 4 page tasks, spikes, bugs, enhancements, chores.

When should I use Create Issue?

Create Issue fits situations like: asks to create/file/open an issue; turn a plan task (e.g.

How do I install Create Issue in Claude Code?

Run `npx skills add foyzulkarim/claude-lens --skill create-issue -a claude-code`. Or copy the skill folder (.claude/skills/create-issue in foyzulkarim/claude-lens) into .claude/skills/create-issue in your project. Claude Code loads it when a task matches its description.

How do I install Create Issue in Codex?

Run `npx skills add foyzulkarim/claude-lens --skill create-issue -a codex`. Or copy the skill folder (.claude/skills/create-issue in foyzulkarim/claude-lens) into .agents/skills/create-issue in your project. Codex loads it when a task matches its description.

Can I use Create Issue 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 foyzulkarim/claude-lens --skill create-issue -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-issue, .gemini/skills/create-issue, .github/skills/create-issue and .opencode/skills/create-issue in your project.

What does Create Issue need to run?

Going by SKILL.md and its folder, Create Issue needs a shell for the scripts in its folder and the command-line tools its instructions call (gh and bash). Our summary lists: A Bash shell.

Does Create Issue access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Create Issue 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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Create Issue use?

Create Issue is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create Issue use?

About 2.4k tokens (SKILL.md is roughly 9.6k 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 Create Issue?

Skills that share tags, products or a category with Create Issue: CCPM Project Management (automazeio/ccpm, 8.4k stars), Release Validation (Mesh-LLM/mesh-llm, 3.5k stars), Final Release Review (openai/openai-agents-python, 30k stars) and Ouroboros PM Interview (Q00/ouroboros, 6.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create Issue?

foyzulkarim (a GitHub user) maintains it in foyzulkarim/claude-lens, which has 250 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on August 5, 2026.

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