plan-backlog

plan-backlog is a skill for Claude Code, Codex from scharissis/polako. It costs 0 tokens per session (3,552 once invoked), scanned A, original, MIT.

A planning tool that reads a vision document and a repository, then proposes large work areas and smaller GitHub issues. The issues are marked for human review and are not automatically treated as ready work.

In plain words
What is it for?
Use it to create a reviewed backlog of proposed features, group related issues, and record dependencies between tasks.
Why use it?
It turns a broad product idea into an organized list of possible tasks without changing code or existing project discussions.

Skill for Claude CodeCodex

Part of the polako plugin — 3 skills shipped together

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/scharissis/polako/plan-backlog
Any agent
npx skills add scharissis/polako --skill plan-backlog
Clone the repo
git clone --depth 1 https://github.com/scharissis/polako

Made for: Claude Code, Codex.

Or install polako, the plugin that ships this one along with the rest of its 3 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for plan-backlog

README.md
[![agentmods](https://agentmods.dev/badge/skills/scharissis/polako/plan-backlog.svg)](https://agentmods.dev/skills/scharissis/polako/plan-backlog)
Your own site
<a href="https://agentmods.dev/skills/scharissis/polako/plan-backlog"><img src="https://agentmods.dev/badge/skills/scharissis/polako/plan-backlog.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,552 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 $0.00000 $0.03552
Opus 5 $0.00000 $0.01776
Sonnet 5 $0.00000 $0.00710
Haiku 4.5 $0.00000 $0.00355

Measured yesterday against content hash 3908b166ec36, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-backlog 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 yesterday.

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.

skills/plan-backlog/SKILL.md · 273 lines

How it starts

The opening of the file, as written. The whole thing — 273 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Propose a backlog from $vision

Focus, if one was given: $focus

What this run may do, and what it may not

It reads a repository and files proposals: issues carrying the proposed label, which no unattended run will ever pick up. A human takes that label off, and that act — not this run — is what queues work.

So the whole of the write surface is creating labelled issues. This run does not commit, push, open pull requests, alter existing issues or comments, or touch a line of code. If something here seems to call for one of those, the answer is that it belongs in a proposal for a human to approve, not in this run.

The whole of the gh surface this run may use, today, is three call shapes: the two issue list reads in Phase 0, and the one issue create in Phase 4 — the last of which may carry --parent and --blocked-by, and no other write flag. Nothing else, including anything that looks like diligence rather than a write: do not run gh label list to confirm the proposed label exists, or gh --version, or any other pre-flight probe. Assume the label exists — it is the label this skill mints every issue with. The allowlist is the binary's to grant, not this skill's to widen: polako plan grants this run its tools, polako health grants its twin review-health only issue list and issue create, and an operator's -tools can narrow either verb further. issue list and issue create are the floor every shipped verb grants, so they are the only surface this skill can rely on — a call outside them raises a permission prompt nobody is there to answer under an unattended drain, and the run hangs instead of reporting. Improvising one, however reasonable it looks, is worse than skipping a check that was never asked for.

The document is required

If $vision is empty, stop and ask which document to work from. Never guess which file is the roadmap: picking the wrong one produces a plausible backlog for a project nobody is building, and the mistake is invisible until a curator reads seven issues that make no sense.

Read the full file on GitHub · 273 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. yesterday Changed · +32 lines 3908b166ec36
  2. 4d ago First seen · 241 lines · 0 tokens per session scan A 91ad68fd31b7

Subscribe to this mod's changes

plan-backlog is a skill published in the GitHub repository scharissis/polako (1 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,552 tokens. 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.

Related

Other skills, from other repositories

pm-portfolio-v1

Portfolio and roadmap batch intake, turn a pile of asks into a prioritised verdict. Use when someone hands a BATCH and wants a decision: "prioritise this backlog", "score these requests", "batch verdict", "what goes in the next sprint or quarter", "turn this backlog document into a roadmap", "sprint recommendation"…

naderelewa/Product-to-Prod · 166 tokens

compound

Document solved problems to compound team knowledge. Creates categorized docs with YAML frontmatter. Triggers on "that worked", "it's fixed", "compound".

wsauret/flywheel · 32 tokens

research

Conduct comprehensive codebase research producing a persistent document. Use when you need to understand something BEFORE planning, or for pure exploration. Triggers on "research", "investigate", "explore codebase".

wsauret/flywheel · 43 tokens

plan-eng-review

Engineering plan review. Reviews a plan document for architecture, code quality, test coverage, and performance before implementation begins. Produces a coverage diagram, failure modes map, and parallelization strategy. Use when asked to "review this plan", "eng review", or "architecture review".

HECer/yoke · 61 tokens

authoring-prd

Use when turning a product idea or change into a loop-ready continuous backlog with small stories and executable behavioral evidence.

HECer/yoke · 27 tokens

writing-for-agents

Create or edit agent-facing instructions such as AGENTS.md, CLAUDE.md, skills, roles, and workflow documents. Use when triggers, completion criteria, context pointers, instruction hierarchy, or duplication affect whether an agent can execute the document reliably.

HECer/yoke · 55 tokens