Repository navigation
Releases: 9gag/openspec-viewer
Releases · 9gag/openspec-viewer
Release list
v0.3.9
v0.3.8
v0.3.7
merge: a requirement's paragraphs are paragraphs The two mains carry the same work under different commits, so this brings across the one change this one does not have: the spec renderer's paragraph override asks Astryx Text for block display, so the labelled paragraphs of a requirement sit apart with the spacing the artifact styles already gave them, instead of running together as one inline sentence. One file, and nothing on this side had touched it since the last merge, so there was no overlap. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
v0.3.6
merge: a capability exists from the first document filed in it The two mains carry the same work under different commits, so this brings across the one change this one does not have: inside a change, a capability directory is read from the first document filed in it rather than from the spec that arrives last, so an artifact written before the deltas is no longer reported missing. The only overlap was the line of comment naming the walk in board.mjs. The conflicts card and everything reading the store at main are untouched. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
v0.3.4
v0.3.3
v0.3.2
v0.3.1
chore: a review skill that reads the diff on two axes Ported from a private repo's copy, which could not be taken as it stood: it named that repo's organisation in two places, and this package publishes to a registry where a name from the store the tool is developed against must not appear. Half its content pointed at infrastructure this repo does not have — a planning CLI, a backend workspace, an env package, five convention docs and as many sibling skills. What survives is the part worth having: Standards and Spec as separate axes, run as parallel sub-agents and reported side by side without reranking, so code that follows every convention while implementing the wrong thing cannot pass by looking tidy. The Fowler smell baseline comes across whole, since it is about code rather than about any repo. Retargeted to what this repo actually documents: CONTEXT.md as the language, README's Layout for the server/src seam, openspec/config.yaml for the artifact rules, and the commit skill for the rule about private names — that last one greppable rather than readable, so the brief says to grep. The fixed point defaults to the last release tag rather than main, because work here lands on main and comparing against it would give an empty diff. Earned its place on first use: it found a stale route object dropping a position on a same-route navigation, and a glossary entry forbidding the word its own definition used. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011QrRnKshigF9MqpK7vD4wN
v0.3.0
The library entries: lib/store for the store-side readings — idle claims, conflicts, capability state — and lib/spec for parsing a spec's own shapes. Both are published surfaces the package now owes a consumer, which is why this is a minor rather than a patch. Also carries Astryx 0.5.2, up from 0.1.9.
v0.2.9
Merge branch 'seankcw:main' into main Forty-four commits: the git-read work that took a board poll from 700ms to 70ms and got the CLI spawn off the first page load, namespaces as the organising idea across the nav, the index, the board and a page of their own, three ways to read a spec, and the documents a store files beside its specs. package.json is the only conflict, and the same one every sync here has: the version stays 0.2.5 and the pnpm pin stays, since both are this fork's and neither is upstream's to set. The description and the mermaid dependency come from upstream, which is where they were changed.