@pmelab/gtd 55.0.0 → 56.0.0

package/src/workflows/feature/review-answers.feature

@@ -1,324 +0,0 @@
-@inmem
-Feature: every note in the review is answered with one tick
-
-  Each quality note in `.gtd/REVIEW.md` ends in three boxes — fix as
-  suggested, won't fix, or the human's own answer written over the
-  placeholder. The gate refuses to land until every note holds exactly one
-  tick. A fix goes to the nit batch, a won't-fix is dropped and recorded in
-  the gate's commit, an own answer is judged like any other note. Nothing is
-  deferred to `.gtd/TODO.md`.
-
-  Background:
-    Given the shipped workflows, as gtd install scaffolds them
-    And an environment variable "GTD_PROTOTYPEROUNDS" set to "0"
-    And an environment variable "GTD_QUALITYREVIEWS" set to "correctness"
-    And an environment variable "GTD_FLAGMODE" set to "false"
-    And a file "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      """
-    And the working tree is committed as "feat: add calculator"
-    And a file "TODO.md" with:
-      """
-      # Add subtract
-
-      subtract(a, b) returns the difference.
-      """
-    And step "unwind" changes nothing
-    And step "start-gate.check" changes nothing
-    And step "design.triage" writes ".gtd/REQUIREMENTS.md" with:
-      """
-      # Requirements
-
-      ## Concerns
-
-      ### Subtract
-
-      - `subtract(a, b)` returns `a - b`.
-
-      ## Impact
-
-      - the greeting module
-
-      ## Acceptance Criteria
-
-      - AC-1: subtract(a, b) returns a - b.
-      """
-    And step "design.gate.answer" changes nothing
-    And step "seam.naive" writes "src/calc.test.ts" with:
-      """
-      it("[AC-1] subtracts", () => {})
-      """
-    And step "seam.naive" writes "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      export const subtract = (a: number, b: number) => a - b
-      """
-    And step "seam.check" changes nothing
-    And step "implement.applying" runs its script
-    And step "implement.health.check" changes nothing
-    And step "implement.closing" runs its script
-    And step "build.quality.correctness.reviewing" writes ".gtd/QUALITY.md" with:
-      """
-      # Quality
-
-      ## subtract has no negative test
-
-      Add a test subtracting a larger number.
-
-      ## add could be a method
-
-      A class would read better.
-
-      ## AC-1 ignores floating point
-
-      Should subtract round? The criterion does not say.
-      """
-    And step "build.quality.sorting" runs its script
-    And step "build.review.reviewing" writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-      """
-    And step "build.review.reviewing", when prompted with "- `## AC-1 ignores floating point`", writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      ## AC-1 ignores floating point
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-      """
-    And step "build.review.reviewing", when prompted with "worked on the human's nits", writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      - [ ] ./src/calc.ts#2-2 — now rounds the difference
-      """
-    And step "build.review.carrying" runs its script
-    And judge step "build.review.ranking" answers:
-      """
-      []
-      """
-    And judge step "build.quality.triage" answers:
-      """
-      [{"id":"Q1","answer":"noise","p":0.9},{"id":"Q2","answer":"noise","p":0.9},{"id":"Q3","answer":"design","p":0.9}]
-      """
-    And step "build.review.closing" runs its script
-
-  Scenario: the note ends in its three answer boxes
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    And ".gtd/REVIEW.md" contains:
-      """
-      ## Findings
-
-      ### AC-1 ignores floating point
-
-      _critical: judge unsure_
-
-      Should subtract round? The criterion does not say.
-
-      In the code:
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-
-      Tick exactly one:
-
-      - [ ] fix as suggested
-      - [ ] won't fix
-      - [ ] _your answer_
-      """
-
-  Scenario: a sign-off with the note unanswered is refused
-    When the process is driven
-    Then landing is refused with "answer-completeness: 1 note(s) in .gtd/REVIEW.md not answered"
-    And the process rests at "build.review.await-review"
-
-  Scenario: two ticks on one note are refused
-    When the process is driven
-    And the text "- [ ] fix as suggested" in ".gtd/REVIEW.md" is replaced with "- [x] fix as suggested"
-    And the text "- [ ] won't fix" in ".gtd/REVIEW.md" is replaced with "- [x] won't fix"
-    Then landing is refused with "tick exactly one box per note"
-
-  Scenario: a ticked answer still holding the placeholder is refused
-    When the process is driven
-    And the text "- [ ] _your answer_" in ".gtd/REVIEW.md" is replaced with "- [x] _your answer_"
-    Then landing is refused with "answer-completeness"
-
-  Scenario: won't fix drops the note, records it, and signs off
-    Given judge step "build.review.standards.judge" answers:
-      """
-      [{"id":"standard-1","answer":"no","p":0.9}]
-      """
-    When the process is driven
-    And the text "- [ ] won't fix" in ".gtd/REVIEW.md" is replaced with "- [x] won't fix"
-    And gtd lands
-    And the process is driven
-    Then the process rests at "idle"
-    And the commit "gtd(human): build.review.await-review → build.review.standards.judge" has a body containing "Gtd-Mark: wontfix=AC-1 ignores floating point"
-    And ".gtd/TODO.md" does not exist
-    And no commit body contains "build.review.triage"
-
-  Scenario: fix as suggested goes to the nit batch unjudged, and the note is not asked again
-    Given step "build.review.fix.nits.fixing" writes "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      export const subtract = (a: number, b: number) => Math.round(a - b)
-      """
-    And step "build.review.fix.nits.fixing" is prompted with "Should subtract round?"
-    And step "build.health.check" changes nothing
-    When the process is driven
-    And the text "- [ ] fix as suggested" in ".gtd/REVIEW.md" is replaced with "- [x] fix as suggested"
-    And gtd lands
-    And the process is driven
-    Then the process rests at "build.review.await-review"
-    And no commit body contains "build.review.triage"
-    And ".gtd/REVIEW.md" contains "now rounds the difference"
-    And ".gtd/REVIEW.md" does not contain "AC-1 ignores floating point"
-    And ".gtd/TODO.md" does not exist
-    When gtd lands
-    And the process is driven
-    Then the process rests at "idle"
-
-  Scenario: an own answer is judged like a note
-    Given judge step "build.review.standards.judge" answers:
-      """
-      [{"id":"standard-1","answer":"no","p":0.9}]
-      """
-    And judge step "build.review.triage" answers:
-      """
-      [{"id":"answer-1","answer":"nit","p":0.9}]
-      """
-    And step "build.review.fix.nits.fixing" is prompted with "round half up"
-    And step "build.health.check" changes nothing
-    And step "build.review.fix.nits.fixing" writes "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      export const subtract = (a: number, b: number) => Math.round(a - b)
-      """
-    When the process is driven
-    And the text "- [ ] _your answer_" in ".gtd/REVIEW.md" is replaced with "- [x] round half up"
-    And gtd lands
-    And the process is driven
-    Then the process rests at "build.review.await-review"
-    And the commit subjects end with:
-      """
-      gtd(human): build.review.await-review → build.review.standards.judge
-      gtd(judge): build.review.standards.judge → build.review.triage
-      gtd(judge): build.review.triage → build.review.fix.nits.fixing
-      gtd(agent): build.review.fix.nits.fixing → build.health.check
-      gtd(check): build.health.check → build.review.closing
-      gtd(check): build.review.closing → build.review.reviewing
-      gtd(agent): build.review.reviewing → build.review.ranking
-      gtd(judge): build.review.ranking → build.review.carrying
-      gtd(check): build.review.carrying → build.review.await-review
-      """
-    And ".gtd/REVIEW.md" does not contain "AC-1 ignores floating point"
-
-  Scenario: deleting a note's boxes instead of answering is refused
-    When the process is driven
-    And the text "- [ ] fix as suggested" in ".gtd/REVIEW.md" is replaced with ""
-    And the text "Tick exactly one:" in ".gtd/REVIEW.md" is replaced with ""
-    Then landing is refused with "its answer boxes were deleted"
-
-  Scenario: a review still in the flat layout, every chunk a `## `, is answered by its ticks
-    Given judge step "build.review.standards.judge" answers:
-      """
-      [{"id":"standard-1","answer":"no","p":0.9}]
-      """
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    Given a file ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## AC-1 ignores floating point
-
-      _critical: judge unsure_
-
-      Should subtract round? The criterion does not say.
-
-      In the code:
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-
-      Tick exactly one:
-
-      - [ ] fix as suggested
-      - [x] won't fix
-      - [ ] _your answer_
-
-      ## Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-      """
-    When gtd lands
-    And the process is driven
-    Then the process rests at "idle"
-    And the commit "gtd(human): build.review.await-review → build.review.standards.judge" has a body containing "Gtd-Mark: wontfix=AC-1 ignores floating point"
-    And no commit body contains "build.review.triage"
-
-  Scenario: boxes ticked in the walkthrough answer nothing
-    Given judge step "build.review.standards.judge" answers:
-      """
-      [{"id":"standard-1","answer":"no","p":0.9}]
-      """
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    Given a file ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Findings
-
-      ### AC-1 ignores floating point
-
-      _critical: judge unsure_
-
-      Should subtract round? The criterion does not say.
-
-      In the code:
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-
-      Tick exactly one:
-
-      - [ ] fix as suggested
-      - [x] won't fix
-      - [ ] _your answer_
-
-      ## Walkthrough
-
-      ### Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      Tick exactly one:
-
-      - [x] fix as suggested
-      - [ ] won't fix
-      - [ ] _your answer_
-      """
-    When gtd lands
-    And the process is driven
-    Then the process rests at "idle"
-    And no commit body contains "build.review.fix.nits.fixing"

package/src/workflows/feature/severity-budget.feature

@@ -1,116 +0,0 @@
-@inmem
-Feature: the review's ranking is batched to the judge's budget
-
-  `build.review.ranking` puts as many chunks in one judgment as the budget's
-  even share still holds whole; the rest go to the next. A chunk too big to
-  fit even alone is cut, and a cut chunk is critical.
-
-  Background:
-    Given the shipped workflows, as gtd install scaffolds them
-    And an environment variable "GTD_PROTOTYPEROUNDS" set to "0"
-    And an environment variable "GTD_QUALITYREVIEWS" set to "correctness"
-    And an environment variable "GTD_FLAGMODE" set to "false"
-    And an environment variable "GTD_JUDGEBUDGETBYTES" set to "150"
-    And a file "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      """
-    And the working tree is committed as "feat: add calculator"
-    And a file "TODO.md" with:
-      """
-      # Add subtract
-
-      subtract(a, b) returns the difference.
-      """
-    And step "unwind" changes nothing
-    And step "start-gate.check" changes nothing
-    And step "design.triage" writes ".gtd/REQUIREMENTS.md" with:
-      """
-      # Requirements
-
-      ## Impact
-
-      - the greeting module
-
-      ## Acceptance Criteria
-
-      - AC-1: subtract(a, b) returns a - b.
-      """
-    And step "design.gate.answer" changes nothing
-    And step "seam.naive" writes "src/calc.test.ts" with:
-      """
-      it("[AC-1] subtracts", () => {})
-      """
-    And step "seam.naive" writes "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      export const subtract = (a: number, b: number) => a - b
-      """
-    And step "seam.check" changes nothing
-    And step "implement.applying" runs its script
-    And step "implement.health.check" changes nothing
-    And step "implement.closing" runs its script
-    And step "build.quality.correctness.reviewing" writes ".gtd/QUALITY.md" with:
-      """
-      # Quality
-
-      ## AC-1 ignores floating point
-
-      Should subtract round? The criterion does not say.
-      """
-    And judge step "build.quality.triage" answers:
-      """
-      [{"id":"Q1","answer":"design","p":0.9}]
-      """
-    And step "build.quality.sorting" runs its script
-    And step "build.quality.closing" runs its script
-    And step "build.review.carrying" runs its script
-
-  Scenario: a chunk the budget cuts is listed as critical, the others judged in batches of their own
-    Given step "build.review.reviewing" writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      ## AC-1 ignores floating point
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-      """
-    And judge step "build.review.ranking", when its questions contain "C1.security", answers:
-      """
-      [{"id":"C1.security","answer":"no","p":0.9},{"id":"C1.reversibility","answer":"two-way","p":0.9},{"id":"C1.blastRadius","answer":"local","p":0.9},{"id":"C1.newConcept","answer":"no","p":0.9}]
-      """
-    And judge step "build.review.ranking", when its questions contain "C2.security", answers:
-      """
-      [{"id":"C2.security","answer":"no","p":0.9},{"id":"C2.reversibility","answer":"two-way","p":0.9},{"id":"C2.blastRadius","answer":"local","p":0.9},{"id":"C2.newConcept","answer":"no","p":0.9}]
-      """
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    And ".gtd/REVIEW.md" contains:
-      """
-      ## Findings
-
-      ### AC-1 ignores floating point
-
-      _critical: judge unsure_
-      """
-    And ".gtd/REVIEW.md" contains:
-      """
-      ## Walkthrough
-
-      ### Subtract
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-      """
-    And the commit subjects end with:
-      """
-      gtd(agent): build.review.reviewing → build.review.ranking
-      gtd(judge): build.review.ranking
-      gtd(judge): build.review.ranking → build.review.carrying
-      gtd(check): build.review.carrying → build.review.await-review
-      """

package/src/workflows/feature/severity-sorted-review.feature

@@ -1,198 +0,0 @@
-@inmem
-Feature: the final review lists the findings that matter most first
-
-  Before the review rests, the judge rates every chunk the reviewer wrote on
-  the same four questions as a human's note (`build.review.ranking`): critical
-  (security, or a one-way door), should (a contract, or a new concept), could.
-  The quality findings go under `## Findings` in that order, security before
-  one-way doors, each led by its tier and reason; a finding the judge cannot
-  rate is critical. The walkthrough goes under `## Walkthrough` in the
-  reviewer's order, with no tier.
-
-  Background:
-    Given the shipped workflows, as gtd install scaffolds them
-    And an environment variable "GTD_PROTOTYPEROUNDS" set to "0"
-    And an environment variable "GTD_QUALITYREVIEWS" set to "correctness"
-    And an environment variable "GTD_FLAGMODE" set to "false"
-    And a file "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      """
-    And the working tree is committed as "feat: add calculator"
-    And a file "TODO.md" with:
-      """
-      # Add subtract
-
-      subtract(a, b) returns the difference.
-      """
-    And step "unwind" changes nothing
-    And step "start-gate.check" changes nothing
-    And step "design.triage" writes ".gtd/REQUIREMENTS.md" with:
-      """
-      # Requirements
-
-      ## Impact
-
-      - the greeting module
-
-      ## Acceptance Criteria
-
-      - AC-1: subtract(a, b) returns a - b.
-      """
-    And step "design.gate.answer" changes nothing
-    And step "seam.naive" writes "src/calc.test.ts" with:
-      """
-      it("[AC-1] subtracts", () => {})
-      """
-    And step "seam.naive" writes "src/calc.ts" with:
-      """
-      export const add = (a: number, b: number) => a + b
-      export const subtract = (a: number, b: number) => a - b
-      """
-    And step "seam.check" changes nothing
-    And step "implement.applying" runs its script
-    And step "implement.health.check" changes nothing
-    And step "implement.closing" runs its script
-    And step "build.quality.correctness.reviewing" writes ".gtd/QUALITY.md" with:
-      """
-      # Quality
-
-      ## AC-1 ignores floating point
-
-      Should subtract round? The criterion does not say.
-
-      ## Token in logs
-
-      The caller's token reaches the log.
-      """
-    And judge step "build.quality.triage" answers:
-      """
-      [{"id":"Q1","answer":"design","p":0.9},{"id":"Q2","answer":"design","p":0.9}]
-      """
-    And step "build.quality.sorting" runs its script
-    And step "build.quality.closing" runs its script
-    And step "build.review.carrying" runs its script
-
-  Scenario: findings are sorted by tier, the walkthrough kept in the reviewer's order
-    Given step "build.review.reviewing" writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      The new export.
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      ## AC-1 ignores floating point
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-
-      ## Session token
-
-      - [ ] ./src/calc.ts#1-1 — reads the caller's token
-
-      ## Token in logs
-
-      - [ ] ./src/calc.ts#1-1 — where it is logged
-      """
-    And judge step "build.review.ranking" answers:
-      """
-      [{"id":"C1.security","answer":"no","p":0.9},{"id":"C1.reversibility","answer":"two-way","p":0.9},{"id":"C1.blastRadius","answer":"local","p":0.9},{"id":"C1.newConcept","answer":"no","p":0.9},{"id":"C2.security","answer":"no","p":0.9},{"id":"C2.reversibility","answer":"two-way","p":0.9},{"id":"C2.blastRadius","answer":"contract","p":0.9},{"id":"C2.newConcept","answer":"no","p":0.9},{"id":"C3.security","answer":"yes","p":0.9},{"id":"C3.reversibility","answer":"two-way","p":0.9},{"id":"C3.blastRadius","answer":"module","p":0.9},{"id":"C3.newConcept","answer":"no","p":0.9},{"id":"C4.security","answer":"no","p":0.9},{"id":"C4.reversibility","answer":"one-way","p":0.9},{"id":"C4.blastRadius","answer":"contract","p":0.9},{"id":"C4.newConcept","answer":"no","p":0.9}]
-      """
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    And ".gtd/REVIEW.md" contains:
-      """
-      <!-- base: 0000000 -->
-
-      ## Findings
-
-      ### Token in logs
-
-      _critical: one-way door_
-
-      The caller's token reaches the log.
-      """
-    And ".gtd/REVIEW.md" contains:
-      """
-      - [ ] _your answer_
-
-      ### AC-1 ignores floating point
-
-      _should: contract_
-
-      Should subtract round? The criterion does not say.
-      """
-    And ".gtd/REVIEW.md" contains:
-      """
-      - [ ] _your answer_
-
-      ## Walkthrough
-
-      ### Subtract
-
-      The new export.
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      ### Session token
-
-      - [ ] ./src/calc.ts#1-1 — reads the caller's token
-      """
-    And ".gtd/REVIEW.md" does not contain "_critical: security_"
-    And the commit "gtd(judge): build.review.ranking → build.review.carrying" has a body containing "Gtd-Judge:"
-    And the commit subjects end with:
-      """
-      gtd(agent): build.review.reviewing → build.review.ranking
-      gtd(judge): build.review.ranking → build.review.carrying
-      gtd(check): build.review.carrying → build.review.await-review
-      """
-
-  Scenario: a finding the judge is unsure about is listed as critical
-    Given step "build.review.reviewing" writes ".gtd/REVIEW.md" with:
-      """
-      # Review: abc1234
-
-      <!-- base: 0000000 -->
-
-      ## Subtract
-
-      The new export.
-
-      - [ ] ./src/calc.ts#2-2 — subtract(a, b)
-
-      ## AC-1 ignores floating point
-
-      - [ ] ./src/calc.ts#2-2 — returns the raw difference
-
-      ## Session token
-
-      - [ ] ./src/calc.ts#1-1 — reads the caller's token
-
-      ## Token in logs
-
-      - [ ] ./src/calc.ts#1-1 — where it is logged
-      """
-    And judge step "build.review.ranking" answers:
-      """
-      [{"id":"C1.security","answer":"no","p":0.9},{"id":"C1.reversibility","answer":"two-way","p":0.9},{"id":"C1.blastRadius","answer":"local","p":0.9},{"id":"C1.newConcept","answer":"no","p":0.9},{"id":"C2.security","answer":"no","p":0.4},{"id":"C2.reversibility","answer":"two-way","p":0.9},{"id":"C2.blastRadius","answer":"local","p":0.9},{"id":"C2.newConcept","answer":"no","p":0.9},{"id":"C3.security","answer":"yes","p":0.9},{"id":"C3.reversibility","answer":"two-way","p":0.9},{"id":"C3.blastRadius","answer":"local","p":0.9},{"id":"C3.newConcept","answer":"no","p":0.9},{"id":"C4.security","answer":"no","p":0.9},{"id":"C4.reversibility","answer":"two-way","p":0.9},{"id":"C4.blastRadius","answer":"local","p":0.9},{"id":"C4.newConcept","answer":"yes","p":0.9}]
-      """
-    When the process is driven
-    Then the process rests at "build.review.await-review"
-    And ".gtd/REVIEW.md" contains:
-      """
-      ## Findings
-
-      ### AC-1 ignores floating point
-
-      _critical: judge unsure_
-      """
-    And ".gtd/REVIEW.md" contains:
-      """
-      ### Token in logs
-
-      _should: new concept_
-      """