issue-296: Rewrite extensions/grillme/README.md in STE style and add the Response style section #306

Merged
david merged 2 commits from feature/issue-296/rewrite-extensions-grillme-readme-md-in-ste-style into main 2026-10-09 22:17:00 +00:00
Owner

296: Rewrite extensions/grillme/README.md in STE style and add the Response style section

Closes #296 — 296.

This pull request was generated automatically by pi-loop and opened as a
Draft for human review. Do not merge without reviewing the changes.

Generated in 6m 20s.

Review findings

I'll start by reading the README to see the current state, then apply the three fixes.Now applying the three fixes.Edits look good and sentences stay under the word limits. Running the gates.All three findings fixed in a single pass; only extensions/grillme/README.md changed. Gates re-run after the edits: root npm test → 1468 pass / 1 skip / 0 fail (1469 tests, 31 suites); extensions/grillme node --test → 466 pass / 0 fail; the documented tsc --noEmit … check → clean (exit 0). No lint config exists in the repo.

Review Findings — 296

Summary

All three findings are resolved with minimal edits to extensions/grillme/README.md. The Medium finding was fixed by restoring the shipped actor ("the assistant asks the user to initialize, create, or select a repo or remote") and re-splitting the sentence so every sentence stays under the 20-word instruction limit. The two Low findings were fixed by restoring the disambiguating words "planning" (intro) and "grilling" (Behavior), with paragraphs reflowed to keep the surrounding line-wrapping style. No code was touched. Verification: root npm test → 1468 pass / 1 skip / 0 fail (31 suites); cd extensions/grillme && node --test → 466 pass / 0 fail; documented tsc --noEmit … index.ts bash-filter.ts prompt.ts ste-style.ts widget.ts → clean. All changed sentences measured at ≤14 words, and no inline-code or code-fence tokens were altered.

Critical

High

Medium

  • extensions/grillme/README.md:108 — "It can initialize, create, or select a repo or remote." attributes the action to the assistant, but the shipped guidance is that the assistant asks the user to do it (extensions/grillme/prompt.ts:96: "ask the user to initialize, create, or select one", reused verbatim by index.ts:754 / index.ts:779). The previous README read "the assistant asks to initialize, create, or select a repo/remote", so this rewrite changes the actor and contradicts the "write the README against the shipped behaviour" requirement. A user could wrongly expect the assistant to create/initialise the repo itself. Suggested fix: reword to "The assistant asks the user to initialize, create, or select a repo or remote."

Low

  • extensions/grillme/README.md:3 — The intro now says "thorough Socratic sessions", dropping "planning". That under-describes the extension and disagrees with its own package metadata (extensions/grillme/package.json description: "Socratic planning and shared-understanding sessions for pi") and the root README.md table entry ("Socratic planning/grilling sessions"). Suggested fix: restore the word, e.g. "Pi Grill Me is a pi extension for thorough Socratic planning sessions."

  • extensions/grillme/README.md:46 — "Grill Me uses one thorough default style." now collides with the new ## Response style section, so "style" carries two meanings in one document (grilling style vs. writing style), against rule 6 ("Use one word for one meaning"). This word was "grilling" before the rewrite. Suggested fix: restore the disambiguator, e.g. "Grill Me uses one thorough default grilling style."

## 296: Rewrite extensions/grillme/README.md in STE style and add the Response style section Closes #296 — [296](https://git.excelera.net/david/pi-extensions-and-skills/issues/296). > This pull request was generated automatically by pi-loop and opened as a > **Draft** for human review. Do not merge without reviewing the changes. Generated in 6m 20s. ## Review findings I'll start by reading the README to see the current state, then apply the three fixes.Now applying the three fixes.Edits look good and sentences stay under the word limits. Running the gates.All three findings fixed in a single pass; only `extensions/grillme/README.md` changed. Gates re-run after the edits: root `npm test` → 1468 pass / 1 skip / 0 fail (1469 tests, 31 suites); `extensions/grillme` `node --test` → 466 pass / 0 fail; the documented `tsc --noEmit …` check → clean (exit 0). No lint config exists in the repo. # Review Findings — 296 ## Summary All three findings are resolved with minimal edits to `extensions/grillme/README.md`. The Medium finding was fixed by restoring the shipped actor ("the assistant asks the user to initialize, create, or select a repo or remote") and re-splitting the sentence so every sentence stays under the 20-word instruction limit. The two Low findings were fixed by restoring the disambiguating words "planning" (intro) and "grilling" (Behavior), with paragraphs reflowed to keep the surrounding line-wrapping style. No code was touched. Verification: root `npm test` → 1468 pass / 1 skip / 0 fail (31 suites); `cd extensions/grillme && node --test` → 466 pass / 0 fail; documented `tsc --noEmit … index.ts bash-filter.ts prompt.ts ste-style.ts widget.ts` → clean. All changed sentences measured at ≤14 words, and no inline-code or code-fence tokens were altered. ## Critical ## High ## Medium - [x] `extensions/grillme/README.md:108` — "It can initialize, create, or select a repo or remote." attributes the action to the assistant, but the shipped guidance is that the assistant *asks the user* to do it (`extensions/grillme/prompt.ts:96`: "ask the user to initialize, create, or select one", reused verbatim by `index.ts:754` / `index.ts:779`). The previous README read "the assistant asks to initialize, create, or select a repo/remote", so this rewrite changes the actor and contradicts the "write the README against the shipped behaviour" requirement. A user could wrongly expect the assistant to create/initialise the repo itself. Suggested fix: reword to "The assistant asks the user to initialize, create, or select a repo or remote." ## Low - [x] `extensions/grillme/README.md:3` — The intro now says "thorough Socratic sessions", dropping "planning". That under-describes the extension and disagrees with its own package metadata (`extensions/grillme/package.json` description: "Socratic planning and shared-understanding sessions for pi") and the root `README.md` table entry ("Socratic planning/grilling sessions"). Suggested fix: restore the word, e.g. "Pi Grill Me is a pi extension for thorough Socratic planning sessions." - [x] `extensions/grillme/README.md:46` — "Grill Me uses one thorough default style." now collides with the new `## Response style` section, so "style" carries two meanings in one document (grilling style vs. writing style), against rule 6 ("Use one word for one meaning"). This word was "grilling" before the rewrite. Suggested fix: restore the disambiguator, e.g. "Grill Me uses one thorough default grilling style."
david merged commit 43b599e163 into main 2026-10-09 22:17:00 +00:00
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
david/pi-extensions-and-skills!306
No description provided.