Repository navigation
Description
Scheduled dependency check against origin/master @ 61672aa (2026-10-05). Cross-checked every pin in nvim-pack-lock.json via git ls-remote + commit-range review; only genuinely new findings (not already covered by #327, #273, #271, #255, #154, #101, #138, #78) are below.
mini.nvim — breaking change upstream, pin not yet affected but worth planning for
Pinned 561751e (2026-09-22) vs upstream HEAD 65d615e (2026-10-04), 11 commits. Among them:
c0a9015 fix(ALL)!: remove compatibility code for outdated features— breaking (conventional-commit!). Message: "All new features were released fairly long ago with compatibility layer for previous behavior... hopefully everybody interested migrated as expected."db75dd4 fix(surround): prepare to drop use_nvim_treesitter option436b4fd fix(ai): prepare to drop use_nvim_treesitter option
This config's mini.ai.setup() / mini.surround.setup({...}) calls (lua/config/plugin_config.lua:15-40) don't set use_nvim_treesitter explicitly, so the current defaults aren't immediately broken by this range. But the ! commit removes back-compat shims outright, and the use_nvim_treesitter option is now explicitly slated for removal — worth a deliberate review (not a trivial bump) before next pulling mini.nvim past 65d615e. Not opening a PR for this one.
Corrections to existing open issues (ground-truth re-check)
- [deps] nvim-treesitter-textobjects pinned to frozen
masterbranch — migrate tomain#101 (nvim-treesitter-textobjects pinned to frozenmaster): the repo's remoteHEADsymref now points atrefs/heads/main, and the current pinned rev (5c7b0263) is identical tomain's current tip. There is no drift and nothing to migrate —vim.pack's default-branch clone is already trackingmain, not the frozenmasterbranch. [deps] nvim-treesitter-textobjects pinned to frozenmasterbranch — migrate tomain#101's premise appears to no longer hold (same pattern as the stale [deps] nvim-treesitter repository archived (read-only since April 2026) — migration path needed #255 archival claim); worth a comment there rather than a new migration PR. - [deps] diffview.nvim abandoned since 2024 — maintained fork available; quicker.nvim v1.5.1 unchecked #273 (quicker.nvim v1.5.1 unchecked): confirmed — the current pin (
4a6883c) isv1.5.1(also taggedstableupstream). Already current, nothing to bump. - [deps] gitsigns.nvim v1.0.0 + v2.0.0 breaking changes — review before next SyncPkgs #154 (gitsigns.nvim v1/v2 breaking changes): upstream has since tagged
v2.1.0(a462f41, past thev2.0.0already flagged); our pin (070a5d7) is still pre-v1.0.0. No new action, just noting the gap has grown. - [deps] fff.nvim at 0.9.6 (config targets 0.9.4); blink.cmp v2 imminent with breaking config changes #271/[audit] fix: bound blink.cmp version range to stay off v2 #328 (blink.cmp v2 risk from an "open" version pin): re-verified the semantics —
vim.version.range("1.10.0")(a fully-qualified x.y.z string) produces an exact range (from == to == 1.10.0), matching whatnvim-pack-lock.jsonitself records ("version": "1.10.0 - 1.10.0"). This is a closed, single-version pin, not an open/unbounded one as [audit] fix: bound blink.cmp version range to stay off v2 #328 assumed — which is consistent with [audit] fix: bound blink.cmp version range to stay off v2 #328 having been closed without merging. The pin already fully blocks ablink.cmpv2 install; [deps] fff.nvim at 0.9.6 (config targets 0.9.4); blink.cmp v2 imminent with breaking config changes #271's recommendation to plan the eventual v2 migration still stands, just not as an urgent pin-safety gap. Side note: upstream has since released bugfix patchesv1.10.1/v1.10.2(both additive/bugfix per commit log, no breaking markers) that the exact pin excludes — low priority, mention only.
Checked, no drift, no action
blink.lib, sneaks.vim, guh.nvim, fugitive.vim (verified via github.com/tpope/vim-fugitive mirror since tpope.io is blocked from this environment), diffview.nvim, async.nvim, refactoring.nvim, conform.nvim, nvim-lint, tiny-inline-diagnostic.nvim, obsidian.nvim, render-markdown.nvim, plenary.nvim, snacks.nvim, noice.nvim, nui.nvim, fx.nvim, fff.nvim (within its 0.9.4 pin, already covered by #271) all have pinned rev == current upstream tip.
A separate, trivial-bump PR for fzf-lua + nvim-lspconfig (both additive/docs-only drift) was opened alongside this issue.
🤖 Generated with Claude Code
Activity