feat: groundtruth doctor + fact-checked launch kit (v0.7.0)#29
Merged
Conversation
Adds the highest-ROI onboarding feature and a complete, fact-checked go-to-market kit (produced + adversarially verified by a multi-agent workflow; every claim checked against the codebase and primary research). - feat: `groundtruth doctor` — diagnose an install in one command (Node, hook wiring project+global, binary, git, config, transcript discoverability, verify-loop state, ledger) with actionable hints; exit 1 on any failure. New src/doctor.ts (pure buildDoctorReport seam + scanSettings) + 9 tests. - docs: launch kit under docs/launch/ (Show HN, X thread, Reddit, Product Hunt, dev.to, awesome-claude-code submission, good-first-issues, ordered CHECKLIST) — drafts for a human to post, nothing auto-published. - docs: new docs/comparison.md (vs tests / manual review / AI code reviewers). - docs: README "Why" now cites the research it's built on — Gong et al., MSR'26 (arXiv:2601.04886): never-implemented changes are the veltiq#1 message-vs-code inconsistency (45.4%), accepted 51.7% less often. Citation independently verified against the paper before landing. - chore: refreshed assets/social-card.svg (new tagline + the catch); plugin.json synced to 0.7.0. 150 tests pass, typecheck + biome clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two things: the highest-ROI onboarding feature, and a complete go-to-market kit (every claim fact-checked against the codebase and primary research — non-negotiable for an anti-hallucination tool).
groundtruth doctorOne command answers "is it actually wired up and working?" — checks Node, the Stop hook (project + global), the binary, git evidence, config, transcript discoverability, the verify-loop state, and the ledger, with an actionable hint when anything's off (exit 1 on failure).
New
src/doctor.ts(purebuildDoctorReport+scanSettingsseam) + 9 tests.Launch kit (
docs/launch/)Fact-checked drafts for a human to post (nothing auto-published): Show HN, X thread, Reddit, Product Hunt, dev.to, the awesome-claude-code submission, good-first-issues, and an ordered CHECKLIST. Plus
docs/comparison.md(vs tests / manual review / AI code reviewers) and a refreshed social card.Produced and adversarially fact-checked by a multi-agent workflow; one fabricated stat was caught and removed, and the README "Why" now cites the real research it was built on — Gong et al., MSR'26 (arXiv:2601.04886): never-implemented changes are the #1 message-vs-code inconsistency (45.4%), accepted 51.7% less often. Citation independently verified against the paper before landing.
Quality
150 tests pass (+9), typecheck + biome clean.
doctorbehaviorally smoke-tested (wired vs unwired, exit codes). Ships as 0.7.0.🤖 Generated with Claude Code