answer-plainly

A preserved method from https://github.com/freddie-northam/skills at e8417d4e7724, path skills/answer-plainly, MIT. 27 of 45 source lines differ (60%), every difference claimed by an entry of the ledger with its reason. Entries: baseline-copies-2026-09-11, fold-2026-09-11.

dependency 1
harness 1
scope 2

Files

SKILL.md 4

Every difference, as it stands

`SKILL.md`

---

harnesschangedbaseline-copies-2026-09-11

greenline renders its own frontmatter: quoted name and description, the description from the manifest override where one existed, no upstream activation flag

description: >-

description: "Use when the user asks a direct question about a codebase, a result, a status, or a risk; when reporting what you did; and whenever you are about to write \"significantly\", \"much faster\", \"should work\", \"generally\", or \"nearly all\"."

Use when the user asks a direct question about a codebase, a result, a status,

or a risk; when reporting what you did; and whenever you are about to write

"significantly", "much faster", "should work", "generally", or "nearly all".

---

# Answer plainly

scopechangedfold-2026-09-11

The opening paragraph says the discipline governs replies, reports and review findings on the agent's own judgment and is never announced, that a measured number enters an artifact with its claim, that the process is not the answer, that show-me decides the form after the answer, that dense wording goes through unslop, and that a contradicted answer is re-measured through verify-this and never repeated unchanged; the body's four-answer rule is scoped in its own sentence to yes/no, quantity and unknown-result questions, a known name, path, definition or explanation being answered directly with its evidence, so the rule stops forcing a yes/no prefix or invented uncertainty onto factual answers (a QA finding behind the baseline prelude).

A question has four permitted answers.

This discipline governs replies, reports and review findings on your own

judgment, never announced; where a claim enters an artifact, the measured

number goes in with it. The four-answer rule below covers yes/no, quantity and

unknown-result questions. A known name, path, definition or explanation is

answered with that answer itself and the repository evidence behind it, with

no yes/no prefix and no invented uncertainty; a number stays required for a

quantitative claim, not for every fact. The process is not the answer: report

what is true, not the tools you are about to run, unless the user asked how

you work. Answer first, then the view: show-me decides what form the answer

takes, after the answer. Dense wording goes through unslop. When the user

contradicts an answer you already gave, the reply is a fresh measurement,

never the same sentence again: re-run the check through verify-this, quote

the new verdict, and either name the benign difference that explains their

result or correct the record plainly.

scopechangedfold-2026-09-11

A yes/no, quantity or unknown-result question has four permitted answers.

1. **Yes.**

2. **No.**

3. **A number.**

4 unchanged lines

4. **I do not know**, and what you will do to find out.

Anything else is a paragraph standing in for an answer. Lead with one of the

four, then explain if explanation helps.

## Replace the adjective with the measurement

dependencychangedfold-2026-09-11

The measurement rule names verify-this, the roster stage that records the verdict a quantitative claim quotes, so the number has an owner in the installed roster.

An adjective is a claim you did not measure.

An adjective is a claim you did not measure. verify-this supplies the

measurement: quote its recorded verdict, never an adjective in its place.

| Instead of | Write |

| --- | --- |

16 unchanged lines

| much faster | p90 fell from 10ms to 1ms |

| nearly all users | 87% of accounts |

| significantly better | 25 basis points |

| a large test suite | 9,134 tests |

If you cannot supply the number, the honest answer is the fourth one.

## Never

- "should work", "generally", "in most cases", when you have not checked

- a confident answer to a question the material cannot settle

- an adjective where a count was available

## It's working if

- A yes-or-no question gets a yes or a no in the first sentence.

- A how-many question gets a digit.

- An unknowable question gets "I do not know" rather than a hedge.

- Claims about size, speed or proportion carry numbers.

The timeline

Each entry that touched this method, with the differences it claimed as they stood at its commit, read from the repository's history.

2026-09-11 `baseline-copies-2026-09-11`

the cutover to an edited copy (ADR 0039): the composed output written as the copy, every difference from upstream claimed with the reason of the overlay that produced it

`SKILL.md`

harnesschangedbaseline-copies-2026-09-11

greenline renders its own frontmatter: quoted name and description, the description from the manifest override where one existed, no upstream activation flag

description: >-

Use when the user asks a direct question about a codebase, a result, a status,

or a risk; when reporting what you did; and whenever you are about to write

"significantly", "much faster", "should work", "generally", or "nearly all".

lifecyclechangedbaseline-copies-2026-09-11

greenline prelude, to fold: scopes the four-answer rule to yes/no, quantity and unknown-result questions; routes numbers to verify-this, form to show-me, dense wording to unslop

**greenline prelude: a guideline, not a stage.** Governs replies, reports, and review findings; where a claim enters an artifact, the measured number goes in with it. `verify-this` supplies the number this skill demands: quote its recorded verdict, never an adjective in its place. `show-me` decides what form the answer takes, and it comes after the answer: answer first, then the view. The process is not the answer: report what is true, not the tools you are about to run, unless the user asked how you work. For dense wording, use `unslop` to make the explanation clear. For a fact the evidence cannot settle, say you do not know and name what would settle it. And when the user contradicts an answer you already gave, the reply is a fresh measurement, never the same sentence again: re-run the check through `verify-this`, quote the new verdict, and either name the benign difference that explains their result or correct the record plainly. Repeating the first answer unchanged is the one move that is never right.

**Scope of the four-answer rule.** The body's restriction to four answer forms

applies to yes/no, quantity, and unknown-result questions. For a known name,

path, definition, or explanation, lead with that answer directly and its

supporting repository evidence where needed. This replaces “Anything else”

and “Lead with one of the four” for those questions. A factual answer needs

neither an irrelevant yes/no prefix nor invented uncertainty. Numerical

measurement remains required for quantitative claims; it is not the format

of every fact.

2026-09-11 `fold-2026-09-11`

The two prelude paragraphs are folded into an opening scope paragraph, the scoped four-answer sentence and a verify-this pointer at the measurement rule; a discipline, so no closing section.