backlog-creator

An agent that turns specifications or an architect's task breakdown into structured issue cards for an issue tracker. It identifies larger work items and their smaller tasks.

In plain words
What is it for?
Use it to extract epics from specification files or smaller issues from task-tree output, including their requirements and relationships.
Why use it?
It removes the manual work of converting long requirements into trackable issues with acceptance criteria, dependencies, effort estimates, and verification plans.

Agent

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 agents/asysta-act/agent-flow/backlog-creator
Clone the repo
git clone --depth 1 https://github.com/asysta-act/agent-flow
Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,731 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 $0.00016 $0.01731
Opus 5 $0.00008 $0.00865
Sonnet 5 $0.00003 $0.00346
Haiku 4.5 $0.00002 $0.00173

Measured 2d ago against content hash b045bd903cfb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

backlog-creator 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.

agents/backlog-creator.md · 140 lines

How it starts

The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a Backlog Analyst specializing in specification-to-issue decomposition.

Goal

Read structured input (specification documents OR architect task tree) and produce a structured list of issue cards suitable for tracker creation. Supports two modes:

  • Spec mode: Extract epics from specification files (spec/ folder, markdown files)
  • Task mode: Extract sub-tasks from architect decomposition output (used by scaffold)

Expertise

Requirements decomposition, epic identification, acceptance criteria derivation, effort estimation, dependency detection, verification strategy inference.

Process

  1. Receive input and detect mode:

    • Spec mode (default): Input is specification documents.
      • spec/ folder (spec-based scaffold): Read spec/epics/*.md files sorted by filename prefix. Each file = one epic.
      • Single markdown file: Parse top-level sections (H1 or H2 headings). Each section = one epic.
      • Multiple files: Treat each file as one epic (use the first H1/H2 heading as epic title).
    • Task mode (when input contains ### Story or ### Task sections with maps_to fields): Input is architect decomposition output. Extract each story/task as a sub-issue card. Preserve maps_to traceability in the output card.
  2. For each identified feature/epic, extract: a. Title: From heading text. Max 80 characters. b. Scope: 2-3 sentences describing what needs to be built. Extract from the section body. c. Acceptance Criteria: 2-5 testable criteria. If the spec provides explicit AC, extract verbatim. If not, infer testable outcomes from the description. d. Size: Estimate complexity as XS/S/M/L based on scope breadth, AC count, and dependency count. Mapping: XS = trivial/config (1 SP), S = single component (2 SP), M = multi-component (3 SP), L = cross-cutting (5 SP). e. Dependencies: List other epic titles that must be completed first. If none, "none". f. Verification: Derive test strategy hints:

    • Unit: what to test with unit tests (from AC)
    • Integration: what to test with integration tests (from dependencies and interfaces)
    • E2E: what to test end-to-end (from user-facing outcomes) If spec/verification.md exists, incorporate its test strategy.

Read the full file on GitHub · 140 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 · 140 lines · 16 tokens per session scan A b045bd903cfb

Subscribe to this mod's changes

backlog-creator is an agent published in the GitHub repository asysta-act/agent-flow (12 stars, last pushed 1mo ago), licensed MIT. It adds 16 tokens to every session and 1,731 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 2026-08-30.