mirror of
https://github.com/openprose/prose.git
synced 2026-09-19 05:55:05 +08:00
irl-dan
4bc41a93bb
Get the Reactor packages mergeable + publishable and the launch surfaces coherent. CI / publish (ci-reactor-package.yml): - Drop the deleted reactor-cradle from the workflow and remove the obsolete .mjs release-gate suite (.github/scripts/*reactor*.mjs + tests) entirely. - Add per-package pack + publish jobs for @openprose/reactor, @openprose/reactor-cli, and @openprose/reactor-devtools (idempotent, version-from-package.json, reactor-v* tag-gated, provenance). The two CLI packages now have a release path. - Fix the dangling cradle reference in tools/cli build-release-tarball.mjs; regenerate pnpm-lock.yaml. Install / first-run: - reactor-cli: @openprose/reactor moved peer -> dependency (a global install now pulls the SDK); @openai/agents + zod become optional peers (keyless install stays light). - SDK README: replace non-existent *V0 imports with the real exports (verifyReceipt / LedgerReceipt / inspectReceiptProof / projectReceiptProof). - CLI README + http-server comment: correct the trigger route to POST /trigger/<node>. Front-door + taxonomy: - Promote the Reactor-forward launch README; add the technical report at docs/reactor/v0.1/report.md (replacing the stale v0.1). - Migrate SKILL.md and all packages/std + packages/co contracts off the retired taxonomy (kind: service -> function, kind: system deleted, ### Ensures -> ### Maintains, ### Criteria/### Services folded) onto the Reactor model. Prune the dead cradle-importing examples: drop flat-tokens and release-readiness/reactor-package-example (keeping release-readiness/src, which is tested). Gates: build, test:reactor:offline, test:examples (10/0), test:skill (252/252) all green offline. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2.3 KiB
2.3 KiB
name, kind
| name | kind |
|---|---|
| email-renderer | function |
Shape
self:- populate named slots in an email shell by generating json-render specs from structured report data
- using only components from the provided catalog
delegates: noneprohibited:- writing raw HTML — all output is json-render specs; modifying the email shell — the shell is fixed infrastructure; using components not in the catalog — specs must reference only cataloged components; re-analyzing data or modifying report content — you decide slot content
- you do not re-interpret the data
Parameters
- report: structured report data to render (domain-specific, opaque to this contract)
- slot_definitions: list of named slots the email shell exposes (e.g.,
summary,detail,footer_cta) - slot_catalog: reference to the component catalog — defines which json-render components are available and their accepted props
Returns
- slots: dict of slot_name → json-render spec, one entry per slot defined in slot_definitions. The returned dict has exactly one entry per slot defined in slot_definitions; every spec is a valid json-render object (
{ root, elements }) referencing only catalog components; and slots adapt to report content — empty or irrelevant sections get appropriate alternative content, not blank space.
Errors
- empty-report: the report contains no data to render
- unknown-slot: a slot_name is not present in slot_definitions
- unknown-component: a spec references a component not in the catalog
Invariants
- the email shell structure is never modified — only slot contents are produced
- output is pure json-render specs — never raw HTML, never inline styles, never layout markup
Strategies
- generate a json-render spec per slot — each spec is a
{ root, elements }object where elements reference components from the catalog; the rendering infrastructure handles converting specs to email-safe HTML - adapt slot content to the report — when a section has no relevant data, produce an appropriate alternative (e.g., a placeholder or positive-status message) rather than an empty spec
- keep specs minimal — only include the data each component needs; do not duplicate report fields unnecessarily
- when a report field is missing or null: use the catalog's default/empty-state component for that slot rather than omitting the slot entirely