{{ message }}
Tags: DietrichGebert/ponytail
Tags
fix: align all version manifests to 4.8.1 + guard against drift (#260, … …#262) (#270) * fix: align all version manifests to 4.8.0 + guard against drift (#260, #262) The v4.8.0 release shipped with all four plugin manifests still reading 4.7.0, and both package.json files still at the 0.1.0 npm-init default. So Claude/Codex/Gemini reported 4.7.0 as the latest version (#262) and the project advertised three different versions at once (#260). Bump all six version-bearing files to 4.8.0 so they match the release tag: the four plugin manifests, the root package.json, and ponytail-mcp. Add scripts/check-versions.js, wired into CI, so this cannot recur. It asserts every version file shares one pinned X.Y.Z version, and on a release-tag run that the shared version equals the tag. The existing mutual-agreement check in tests/gemini-extension.test.js could not catch this, because all four manifests were stale at 4.7.0 together. Fixes #260 Refs #262 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix: target 4.8.1 for a clean superseding release v4.8.0 was already tagged with the stale 4.7.0 manifests. Rather than rewrite a published tag, ship the consistent versions as v4.8.1. The CI guard enforces tag == version on the release run. (#260, #262) --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
fix: comprehension-first guard + reuse rung (#245, #217) (#253) #245 "Dangerously lazy": add an operational "fix the root cause, not the symptom" directive — grep every caller of the function you touch and fix the shared function once (the smaller diff). Validated on the agentic benchmark: on a shared-helper bug-fix trap, baseline fixes the root cause 1/6 while ponytail does 6/6 on both Sonnet 4.6 (the model the issue was filed on) and Opus 4.8, verified by reading the produced code. Plain prose ("trace the flow") did not move it; the actionable, lazy-framed directive did. #217 "Missing rung": add ladder rung 2 "Already in this codebase? Reuse it, don't re-write it." Propagated across SKILL.md, AGENTS.md, all agent mirror copies, the hook fallback, and both READMEs (check-rule-copies passes). Benchmark: 4 new deterministic quality-tier tasks (reuse-slug, reuse-money, trace-transfer, trace-amount) with selftest-proven good/bad refs; harness gains multi-file seed support in --selftest, distinctive-behaviour reuse detection, and counts in-file __main__/demo() self-checks as test LOC (not source bloat) for surgical tasks. Full writeup in benchmarks/results/2026-06-22-issue-245-217-comprehension.md. Also carries the in-progress todo-null benchmark task already present in the working tree. Co-authored-by: Dietrich Gebert <dgebert@Dietrichs-MacBook-Pro.local> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
chore: bump version to 4.5.0 (#60) Release-prep bump across all four plugin manifests (Claude Code, Codex, Gemini, Copilot) for v4.5.0. The cross-manifest parity test keeps them aligned. Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
chore: untrack one-off social images committed by mistake (#46) A `git add -A` during the v4.4.0 bump accidentally tracked the announcement art (announce/changelog/ponytail-*.gif) in the repo root. Untrack them and gitignore the pattern; they stay on disk for posting but out of the repo.
PreviousNext
