chore(changelog): land entries as per-change fragments in changelog.d/

[Anarchid]

Problem

The changelog policy is a structural conflict generator: every PR inserts prose at the same textual location (the top of ## Unreleased in CHANGELOG.md), so any PR that outlives another merge conflicts there. Full rationale and design discussion in anima-research/agent-framework#128 — this PR is the chronicle leg of the same ecosystem-wide rollout (same shape as the July policy rollout).

Changes

Entries now land as one uniquely-named file per change — changelog.d/<slug>.<breaking|added|changed|fixed>.md, containing the entry's markdown bullets verbatim (slug = PR number or branch name). Distinct files never conflict in git.

• scripts/release-changelog.mjs (the npm version hook) folds pending fragments plus anything filed directly under ## Unreleased into the new ## X.Y.Z — date section (subsections in Breaking/Added/Changed/Fixed order; fragments join an existing matching subsection, including audience-qualified ### Breaking (on-disk format) headings), deletes the consumed fragments, and keeps all existing refusals. New refusals: unrecognized category suffix and non-bullet fragment content. Chronicle's script was byte-identical to agent-framework's (it is repo-agnostic — reads package.json for name-free version handling), so the new version drops in verbatim.
• changelog.yml soft check now passes on a well-formed new fragment or a CHANGELOG.md edit (transition-friendly for in-flight PRs; no-changelog label escape unchanged). changelog.d/README.md and badly-named fragments do not satisfy it. Chronicle's existing ^src/ path condition (which here matches the Rust crate's src/) is kept as-is.
• publish.yml is untouched — the tag-time section guard and the github-release mirror job both read the assembled versioned section, which exists at tag time exactly as before because assembly runs pre-tag (verified below).
• Migration needs no flag day: direct ## Unreleased edits remain a supported input, merged at release time. (Chronicle's ## Unreleased is currently empty, so nothing even needs carrying over.)
• CONTRIBUTING.md ## Changelog section rewritten; chronicle-specific prose kept verbatim (what-needs-an-entry list: napi surface, store/branch/state semantics, on-disk/wire format, recovery, packaging; the "say plainly whether existing stores keep opening" breaking guidance; the five-target cross-build in the release ritual). Breaking fragments open the bullet naming who needs to act — chronicle's example adapted to - **On-disk format:** …, matching its existing ### Breaking (on-disk format) heading style, which remains available via a direct Unreleased edit.
• PR-template checkbox updated, changelog.d/README.md documents the format in-place (example slug 10-torn-tail.fixed.md), and this PR dogfoods the scheme with its own fragment (changelog-fragments.changed.md), so the next release exercises the pipeline.

Companion to anima-research/agent-framework#128 and siblings in membrane, context-manager, and connectome-host; safe to merge in any order — each repo's scheme is self-contained.

Tests

Release script exercised on copies in a scratch dir through all eight legs (outputs pasted from the runs):

1. Clean cut — 4 fragments (breaking/added/fixed/dogfood-changed) + manual entries filed directly under Unreleased (including an audience-qualified ### Breaking (on-disk format) heading): breaking fragment joined the qualified heading, canonical subsection order, continuation lines preserved, fragments deleted, README.md kept. CHANGELOG.md: released '## 0.4.0 — 2026-08-24' from 4 fragment(s) plus the Unreleased section. Exit 0.
2. Refuse-duplicate (version set to 0.3.0) — a '## 0.3.0' section already exists. Exit 1.
3. Refuse-empty — no fragments, empty Unreleased: nothing to release as 0.4.0 — no fragments in changelog.d/ and no entries under '## Unreleased'. Exit 1.
4. Fragments-only cut (chronicle's real state: empty Unreleased, dogfood fragment only) — clean ### Changed section, fresh empty ## Unreleased kept above, exit 0.
5. Refuse bad category suffix (oops.md) — exit 1 with naming guidance.
6. Refuse non-bullet fragment — exit 1.
7. Tag guard + github-release awk (the exact commands from publish.yml) run on leg 1's assembled output with GITHUB_REF_NAME=v0.4.0: guard passes, awk extracts the 26-line section, non-empty.
8. changelog.yml grep logic simulated on six changed-file sets: src+fragment PASS, src+CHANGELOG PASS (legacy), src-alone FAIL, src+changelog.d/README.md-only FAIL, src+badly-named-fragment FAIL, docs-only PASS.

Compatibility

No format change — docs, CI, and release tooling only; no Rust or napi code touched.

Not verified

A real npm version run end-to-end (the next actual release is the live exercise — same status as the existing tag-time guards, unexercised since #14 landed). The changelog check will run live on this PR itself.

---

• Changelog fragment added — changelog.d/changelog-fragments.changed.md.

🤖 Generated with Claude Code

[Anarchid]

Greptile's review of the membrane companion (antra-tess/membrane#49) found two issues in the shared code, fixed here in 0479dcb: fragment validation now checks every line (top-level prose or a rogue ## heading refuses the release), and the CI check counts only added fragments (--diff-filter=A), so deleting/renaming one no longer satisfies it.

[Anarchid]

A Codex QA pass on the membrane companion (antra-tess/membrane#49, blocking) found two more shared-code issues, fixed here in ab31cdf: a fragment slug taken verbatim from a fix/foo branch name produced a subdirectory the scanner silently skipped (now fail-closed — anything that isn't README.md or a well-formed fragment aborts the release; docs prescribe / → - slugs), and indented ## headings still slipped past the line check (now a strict grammar: two-space continuations, headings/rules refused at any indent). It also asked for automated coverage: scripts/release-changelog.test.mjs runs the real script in a throwaway directory across 18 cases, including every refusal asserting nothing is modified.

[Anarchid]

Round 2 of the Codex QA pass on membrane (antra-tess/membrane#49) — fixed here in 4e104c0: refusals reported via console.error + process.exit(1) lose their stderr when stderr is a pipe on macOS (async pipes; Linux pipes are sync), so refusals now throw and the entry point sets process.exitCode = 1 and returns normally; and the grammar missed spaced thematic breaks (- - -) and headings as bullet content (- ## x), both now refused (24 regression cases).