plan-work

plan-work is a cursor rule for Cursor from danielvm-git/bigpowers. It costs 83 tokens per session (3,360 once invoked), scanned B, original, MIT.

PLANNING SPINE STEP 3 of 3 — Plan the work: write detailed implementation tasks into the active epic capsule (specs/epics/eNN-slug/). Produces countable-story-format .md specs and runnable -tasks.yaml files. Use after slice-tasks (step 2). Not a substitute for scope-work (step 1) or slice-tasks (step 2).

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/danielvm-git/bigpowers/plan-work
Clone the repo
git clone --depth 1 https://github.com/danielvm-git/bigpowers

Made for: Cursor.

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-work

README.md
[![agentmods](https://agentmods.dev/badge/rules/danielvm-git/bigpowers/plan-work.svg)](https://agentmods.dev/rules/danielvm-git/bigpowers/plan-work)
Your own site
<a href="https://agentmods.dev/rules/danielvm-git/bigpowers/plan-work"><img src="https://agentmods.dev/badge/rules/danielvm-git/bigpowers/plan-work.svg" alt="Measured on agentmods" height="20"></a>
Per session 83 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,360 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin unknown 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 $0.00083 $0.03360
Opus 5 $0.00042 $0.01680
Sonnet 5 $0.00017 $0.00672
Haiku 4.5 $0.00008 $0.00336

Measured today against content hash ab732b4d0f44, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

plan-work scanned grade B with 2 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 today.

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.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

2. Add POST /users endpoint → verify: curl -s -X POST http://localhost:3000/users -d '{"email":"[email protected]"}' | jq .id

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

2. Add POST /users endpoint → verify: curl -s -X POST http://localhost:3000/users -d '{"email":"[email protected]"}' | jq .id
.cursor/rules/plan-work.mdc · 272 lines

How it starts

The opening of the file, as written. The whole thing — 272 lines — stays where its author put it; the contents beside it link to each section on GitHub.

story: e45s29

Plan Work

Spine position: Step 3 — scope-work → slice-tasks → plan-work.

Produce a detailed, verifiable implementation plan in the active epic capsule directory (specs/epics/eNN-slug/). Output: a story spec .md file (countable-story-format) and a decoupled eNNsYY-tasks.yaml with runnable verify commands. "I think it works" is not a step.

HARD GATE — Do NOT proceed with a plan until the task's success criteria are clear. If success is ambiguous, convert the task into "step → verify: <cmd>" pairs here before writing tasks — every task ships a runnable verify: or the plan is not done.

RECURSIVE DISCIPLINE — This lifecycle applies to EVERY task, including updating these skills. Never skip planning because a task is "meta" or "just documentation."

Pre-flight

Read: release-plan.yaml, product/SCOPE_LATEST.yaml, active epics/<capsule>/epic.yaml, tech-architecture/tech-stack.md, product/GLOSSARY_LATEST.yaml.

ZOOM-OUT MANDATE (v1.17.0) — If modifying an existing module: (1) State the module's purpose. (2) Name its callers. (3) List its contracts. Cannot answer all three? Stop — scope is misunderstood.

If this plan touches an existing module, run assess-impact first to understand blast radius.

DISCOVERY MANDATE (v1.18.0) — For external API integration, verify the API signature via local docs or search and quote at least one technical detail in the step's context.

MULTIPLE INTERPRETATIONS (HARD GATE) — If the task admits ≥2 valid interpretations, list them and get a user decision before drafting any steps.

COMPLEXITY PUSHBACK (HARD GATE) — Every new abstraction MUST include a one-sentence "Reason for Depth." If it can't be filled non-trivially, the abstraction is premature — use inline code instead.

SLOPCHECK (HARD GATE) — For every external package, tag it [OK], [SUS], or [SLOP]. [SUS]/[SLOP] require human approval before execution.

Invocation modes

Read the full file on GitHub · 272 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. today First seen · 272 lines · 83 tokens per session scan B ab732b4d0f44

Subscribe to this mod's changes

plan-work is a cursor rule published in the GitHub repository danielvm-git/bigpowers (163 stars, last pushed 2d ago), licensed MIT. It adds 83 tokens to every session and 3,360 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.