plan-issue

A planning workflow for turning a GitHub issue—a task or bug report—into a detailed coding plan after examining the repository.

In plain words
What is it for?
Use it to inspect an issue, trace related code and tests, and produce a plan that another developer can review and approve.
Why use it?
It removes the guesswork from deciding which files, classes, and tests need to change before implementation begins.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/alexmond/jhelm/plan-issue
Any agent
npx skills add alexmond/jhelm --skill plan-issue
Clone the repo
git clone --depth 1 https://github.com/alexmond/jhelm

Made for: Claude Code, Codex.

Per session 18 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 688 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00018 $0.00688
Opus 5 $0.00009 $0.00344
Sonnet 5 $0.00004 $0.00138
Haiku 4.5 $0.00002 $0.00069

Measured 2d ago against content hash 6db9a4717e9c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-issue scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/plan-issue/SKILL.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plan Implementation for GitHub Issue #$ARGUMENTS

Produce a detailed, approvable implementation plan before writing any code.


Step 1: Read the issue and announce work started

gh issue view $ARGUMENTS

Extract:

  • Goal / user-facing behaviour
  • Explicit requirements and acceptance criteria
  • Any referenced files, classes, or issues

Then post a comment to announce work has started:

gh issue comment $ARGUMENTS --body "🔧 **Work started** — exploring the codebase and drafting an implementation plan."

Step 2: Explore the codebase

Use Glob and Grep (or the Explore agent for broad searches) to locate:

  1. Files to create — identify the right package and naming convention by finding similar classes
  2. Files to modify — find every call site or injection point that needs updating
  3. Tests to create/update — find the existing test class for each file being modified
  4. Relevant patterns — read 1–2 existing similar classes to match structure (imports, annotations, error handling)

Key places to always check:

  • jhelm-core/src/main/java/org/alexmond/jhelm/core/ — action classes, models, interfaces
  • jhelm-kube/src/main/java/ — Kubernetes client implementation
  • jhelm-cli/src/main/java/ — CLI commands (Picocli)
  • Matching test directories for each module

Step 3: Write the plan

Write the plan to /tmp/plan-$ARGUMENTS.md using this structure:

# Plan: <Issue Title> (Issue #$ARGUMENTS)

## Context
<2–3 sentences on what currently exists and what is missing>

## Files to Create
### 1. `path/to/NewClass.java`
- Purpose and key methods
- Important implementation notes

## Files to Modify
### N. `path/to/ExistingClass.java`
- What changes and why
- Code sketch if helpful

## Test Files to Create/Update
### N+1. `path/to/NewClassTest.java`
- Test cases to cover (success, error, edge cases)

## Verification
```bash
./mvnw test -pl <module>
./mvnw install
./mvnw spring-javaformat:apply && ./mvnw validate

Read the full file on GitHub · 103 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 103 lines · 18 tokens per session scan A 6db9a4717e9c

Subscribe to this mod's changes

plan-issue is a skill published in the GitHub repository alexmond/jhelm (4 stars, last pushed 8d ago), licensed Apache-2.0. It adds 18 tokens to every session and 688 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.