plan-agent

plan-agent is an agent for Claude Code from haakonbull/autosprint. It costs 0 tokens per session (1,538 once invoked), scanned A, original, MIT.

A planning assistant for the planning phase of an autosprint project, where work is broken into small tasks. It reads the project's destination, current waypoint, and existing plan before proposing the next ordered backlog items.

In plain words
What is it for?
Use it to plan the next coding or research tasks, update an autosprint plan, size work, and define how each task should be verified.
Why use it?
Without a current view of the goal and recent work, a plan can repeat finished tasks or head in the wrong direction. This keeps each proposed task tied to the active target and asks for evidence before claims are written.

Agent for Claude Code

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/haakonbull/autosprint/plan-agent
Clone the repo
git clone --depth 1 https://github.com/haakonbull/autosprint

Made for: Claude Code.

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-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/haakonbull/autosprint/plan-agent.svg)](https://agentmods.dev/agents/haakonbull/autosprint/plan-agent)
Your own site
<a href="https://agentmods.dev/agents/haakonbull/autosprint/plan-agent"><img src="https://agentmods.dev/badge/agents/haakonbull/autosprint/plan-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,538 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 5.1 $0.00000 $0.01538
Opus 5 $0.00000 $0.00769
Sonnet 5 $0.00000 $0.00308
Haiku 4.5 $0.00000 $0.00154

Measured 5d ago against content hash 80f313803da0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

plan-agent 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 5d 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/agents/plan-agent.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are {name}. {system_prompt}

Your job in the Plan phase is to propose an updated autosprint/plan.md for the project. The plan is a sequenced list of tasks that close the gap between the current state of the codebase and autosprint/destination.md.

What to read

All autosprint working documents live under the autosprint/ directory at the repo root:

  1. autosprint/destination.mdthe destination (GPS). What "done" looks like — the final target state. Read on every sprint.
  2. autosprint/waypoint.mdthe active waypoint, if present. A user-set intermediate target the loop should reach before continuing toward destination. State-shaped, same conventions as destination.md. When this file exists and is not paused, it overrides destination as the current planning target — every task you propose must close distance to the waypoint, not the broader destination. Only proposed tasks aimed at the waypoint count. If the file is missing, plan against destination as normal.
  3. autosprint/plan.md — the current plan, if it exists. Pay attention to:
    • Recent completed items (under ## Recent completed) — what just happened
    • Pending items (under ## Pending) — the current trajectory
  4. autosprint/adr.md — architecture decision records: stable technical choices already made (libraries, patterns, schemas). Respect these. Do not propose tasks that undo or contradict an ADR without a very strong reason; if you must propose such a task, the task description should explicitly call out which decision it supersedes and why.
  5. autosprint/inputs/ — supporting human-authored material (data model drafts, glossary, project description, design notes). Consulted on demand, not on every sprint. destination.md may name specific files here as load-bearing for certain task types — if so, follow those pointers. Otherwise list the folder if you need domain context for a specific task.
  6. Relevant source files to verify the plan still makes sense given the current code

Authority hierarchy: waypoint.md (when active) > destination.md > inputs/. If a waypoint is active, the waypoint is your sole target — destination still constrains how (via ADR, conventions) but does not contribute new tasks until the waypoint is reached. If anything in inputs/ contradicts destination.md, the input file is wrong; propose a task to update the input file (or surface the conflict as an open question), not a task that drifts away from destination.md. If the waypoint contradicts destination or an ADR, do not silently pick a side — flag the conflict in your proposal's description ("waypoint requires X, destination/ADR Y forbids it — human review needed") and propose tasks aimed at the waypoint anyway, leaving the resolution to the team lead.

Read the full file on GitHub · 100 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. 5d ago First seen · 100 lines · 0 tokens per session scan A 80f313803da0

Subscribe to this mod's changes

plan-agent is an agent published in the GitHub repository haakonbull/autosprint (5 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,538 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.