scope

A task-scoping template that turns a vague coding request into a bounded plan with a goal, included and excluded work, constraints, success criteria, and verification steps.

In plain words
What is it for?
Use it before coding to define the files, behavior, limits, and exact commands or checks that will show the task is done.
Why use it?
It prevents open-ended work from expanding before anyone agrees what should change or how completion will be checked.

Command

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 commands/patrickserrano/lacquer/scope
Clone the repo
git clone --depth 1 https://github.com/patrickserrano/lacquer
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 420 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 $0.00014 $0.00420
Opus 5 $0.00007 $0.00210
Sonnet 5 $0.00003 $0.00084
Haiku 4.5 $0.00001 $0.00042

Measured yesterday against content hash 2425ff66debf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scope 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.

core/commands/scope.md · 36 lines

What it actually says

Turn a vague ask into a bounded, verifiable task BEFORE touching code. Vague, open-ended starts ("where did we leave off", "clean up the UI") are the costliest, lowest-quality sessions — they expand into large exploratory token burns before scope is defined, and they run out of context mid-task. This template forecloses that.

Produce this frame (fill every field; ask the user only for what you genuinely can't infer)

  • Goal — one sentence: the outcome, not the activity.
  • In scope — the specific files/components/behaviors this task touches.
  • Out of scope — what NOT to change (prevents scope creep and context bloat).
  • Constraints — the rules the solution must respect (existing patterns, no new deps, perf/security boundaries, "don't touch X"). This is the field most often left implicit — state it explicitly.
  • Success criteria — how we'll know it's done: the observable end state.
  • Verification — the exact commands/checks that prove it (build, test, lint, a manual run). Name them now so "done" isn't a guess.

Then

  1. Show the frame and get a quick confirm (or proceed if it's unambiguous and low-risk).
  2. Do the work against the frame; if you discover scope was wrong, STOP and re-frame rather than silently widening.
  3. Before declaring done, run the Verification steps and report their actual output.

Resuming? Front-load, don't rebuild

If this is a continuation ("pick up where we left off"), do NOT reconstruct context by re-reading everything — that's the expensive path. Instead: read the last plan/PR/commit, state the ONE next action in a fresh frame above, and start a new session for it rather than extending a long, high-pressure thread.

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 First seen · 36 lines · 14 tokens per session scan A 2425ff66debf

Subscribe to this mod's changes

scope is a command published in the GitHub repository patrickserrano/lacquer (3 stars, last pushed yesterday), licensed MIT. It adds 14 tokens to every session and 420 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 2026-08-31.