taskplane-task-authoring

A set of rules for writing task packets that another automated worker can execute outside the current conversation. The packets use files such as PROMPT.md, STATUS.md, and CONTEXT.md to describe the work and its state.

In plain words
What is it for?
Use it to create task packets, break features into independent work items, record task status and context, and prepare work for autonomous execution.
Why use it?
It makes planned work understandable and runnable in a fresh context without relying on the original discussion. This helps organize, queue, and track larger software tasks.

Cursor rule for Cursor

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 rules/beettlle/pi-spine/taskplane-task-authoring
Clone the repo
git clone --depth 1 https://github.com/beettlle/pi-spine

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 2,426 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 $0.02426
Opus 5 $0.00000 $0.01213
Sonnet 5 $0.00000 $0.00485
Haiku 4.5 $0.00000 $0.00243

Measured 2d ago against content hash c6c866098dcc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

taskplane-task-authoring 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.

.cursor/rules/taskplane-task-authoring.mdc · 216 lines

How it starts

The opening of the file, as written. The whole thing — 216 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Taskplane task authoring

Author runnable task packets for execution outside this chat. Packets are runner-agnostic file contracts: an executor with a fresh context should complete the work using only PROMPT, STATUS, listed docs, and the codebase.

Related: @taskplane-worker-cursor.mdc for implementing a packet in Cursor. Templates: @.cursor/rules/taskplane/prompt-template.md, @.cursor/rules/taskplane/status-template.md.

When this rule applies

Apply when the user asks to:

  • Create, stage, queue, or plan tasks or batches
  • Write or update PROMPT.md, STATUS.md, or task-area CONTEXT.md
  • Break a feature or spec into orchestratable work items

Do not apply for implementation. If the user asks to implement, fix, or ship code for a feature, defer to @taskplane-worker-cursor.mdc unless they also want new packets authored in the same turn—then author first, implement only if explicitly requested after.

If intent is ambiguous (create tasks vs implement now), ask once.

Orchestration artifacts vs repo docs

PROMPT.md, STATUS.md, and area CONTEXT.md under a tasks root are orchestration artifacts, not unsolicited documentation. Creating them when the user requests task work is allowed. See documentation-policy.mdc (Taskplane task packets exception).

Do not create unsolicited README files, change summaries, or PR writeups unless asked.

Packet layout

{tasks-root}/{PREFIX-###-slug}/
├── PROMPT.md     # Definition (immutable above --- divider)
├── STATUS.md     # Execution memory (mirrors PROMPT steps)
├── .reviews/     # Runtime — do not create at author time
└── .DONE         # Runtime — do not create at author time

Optional at area level:

{tasks-root}/CONTEXT.md   # Next Task ID, domain state, tech-debt backlog

Folder naming: PREFIX-###-short-slug (e.g. AUTH-014-add-role-checks).

Before you write

  1. Locate tasks root — Common defaults: taskplane-tasks/, docs/task-management/. Read project config if present (task areas: path, prefix, context file; reference docs; test commands; standards paths; protected docs; never-load list). If unknown, ask the user for tasks-root, ID prefix, and full test command.
  2. Read area CONTEXT.md — Use Next Task ID for the new task; you will increment it in the same edit pass.
  3. Pick the area — Match task domain to the area that owns the primary files being changed. If multiple areas exist and scope is unclear, ask.

Read the full file on GitHub · 216 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 · 216 lines · 0 tokens per session scan A c6c866098dcc

Subscribe to this mod's changes

taskplane-task-authoring is a cursor rule published in the GitHub repository beettlle/pi-spine (3 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,426 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.