Repository navigation
Add init command for OpenSpec - #7
Merged
Merged
Conversation
drvova
pushed a commit
to drvova/VovaSpec
that referenced
this pull request
Oct 13, 2025
Add init command for OpenSpec
shwcsmack
referenced
this pull request
in shwcsmack/c3spec
May 27, 2026
Add the approved Tier 3 planning artifacts for the combined #7/#15 change: brainstorm, proposal, design, delta specs, tasks, and staged implementation plan. Co-authored-by: Cursor <cursoragent@cursor.com>
DumoeDss
referenced
this pull request
in DumoeDss/rasen
Jul 9, 2026
…onical VERIFY VERDICT, tree-fingerprint evidence, never-read-source scoping Closes the gap where verify-change and verify-enhanced computed a verdict but never persisted a machine-checkable evidence artifact, so ship's pre-flight test-skip gate had no canonical verify output to key off, and the verify-family's severity classifications weren't tied to child #1's canonical vocabulary. Also closes a "never read source code" absolute-without-scope gap in the shared QA/DESIGN methodology blocks, and a pre-existing chrome-use parity coverage gap. - WF-1 (evidence file + ship pre-flight wiring): verify-change.ts (both skill and command getters) adds a Save Report step writing rasen/changes/<name>/verification-report.md with the summary scorecard, verdict status line, and grouped findings. - WF-7 (verdict unification, consumes canonical-severity-vocabulary): verify-change maps CRITICAL/WARNING/SUGGESTION and verify-enhanced maps Critical Issues/Warnings/per-stage PASS-FAIL onto canonical Blocker/Major/Minor/Trivial by reference (no re-declaration); per-stage PASS/FAIL kept as a display aid in verify-enhanced. - WF-8 (test-evidence chain): both verify getters emit the machine-checkable "VERIFY VERDICT: <CLEAN|BLOCKED> — Blocker:<n> Major:<n> Minor:<n> Trivial:<n>" status line (CLEAN iff no open Blocker/Major) and record a tree-fingerprinted test-evidence block (command(s) + result + git rev-parse HEAD^{tree}) when they run any test/gate command, matching the review-cycle schema. ship.ts adds verification-report.md to the pre-flight verification-evidence list and names it explicitly in the test-skip gate's evidence sources (skip-gate logic itself unchanged). - SH-1/SH-2 + RV-6/RV-7 (enumerate-and-gate scoping): _shared.ts QA_METHODOLOGY (#5/#7) and DESIGN_METHODOLOGY (#4) "never read source code" absolutes scoped to the exploration/testing/audit phase, with an explicit carve-out naming (a) diff-aware triage and (b) the STANDALONE fix loop — does not reopen child #1's dispatched-mode report-only contract. - D5: chrome-use added to the parity table (functionFactories + GENERATED_SKILL_FACTORIES + both hash maps), closing a pre-existing blind spot noted in the office-hours-dialogue-override review. Installed skills regenerated via node build.js (pnpm build still broken by the in-flight pnpm-workspace.yaml — confirmed identical output). Parity hashes resynced for qa, qa-only, design-review, plus the new chrome-use entries; verified no other template's hash moved. Reviewed CLEAN (review-cycle round 1, non-author confirmed): 1 Minor fixed (verify-change's Save-Report prose falsely claimed /rasen:retro reads verification-report.md — retro.ts's Outcome Artifacts list never enumerates it; trimmed the false claim, /rasen:ship — the real, correctly-wired consumer — remains named), 2 Trivial accepted-known (qa-only carve-out clause (b) references a fix loop qa-only doesn't have, harmless since it only permits; verify-enhanced omits restating "content overrides label" inline, relying on the referenced canonical-severity-vocabulary instead — acceptable by-reference pattern). ship.ts/verify-enhanced.ts absence from the parity registry (they're commands, not templates) is a pre-existing scope gap, not introduced here — relayed to child #3/#5 if in scope there. No whole-file branding-bundling residue this ship: unlike child #1 (which landed mid-rebrand), the rebrand session's openspec->rasen rename is already committed (2ebfae9, ca54b3b), so these 5 files carry no in-flight branding churn — no LEAD ruling needed here. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
BaurinVladislav
pushed a commit
to BaurinVladislav/OpenSpec
that referenced
this pull request
Oct 5, 2026
- D8: clarify isAbsolute check for cross-drive paths (Fission-AI#5) - Task 2.4: add 'usable' to default root selection (Fission-AI#7) - New scenario: discovery skips unusable entry with warning (Fission-AI#7.1) - New scenario: store ID found but folder missing is an error (Fission-AI#7.2) - Clarify list scenario: mark winner for duplicate IDs (Fission-AI#7.3) Refs Fission-AI#1950
BaurinVladislav
pushed a commit
to BaurinVladislav/OpenSpec
that referenced
this pull request
Oct 8, 2026
- D8: clarify isAbsolute check for cross-drive paths (Fission-AI#5) - Task 2.4: add 'usable' to default root selection (Fission-AI#7) - New scenario: discovery skips unusable entry with warning (Fission-AI#7.1) - New scenario: store ID found but folder missing is an error (Fission-AI#7.2) - Clarify list scenario: mark winner for duplicate IDs (Fission-AI#7.3) Refs Fission-AI#1950
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
openspec initcommand to create OpenSpec directory structureChanges
Test Plan