feat(openehr-specs): add subagents and commands (0.2.0)#1
Merged
Conversation
Round out the plugin from skills-only into a full workflow toolkit.
Subagents (agents/) — context-isolated, dispatched by Claude:
- spec-reviewer: runs the review check catalog across a whole spec document
- xref-auditor: verifies {openehr_*} attributes and cross-spec anchors via the
Markdown-twin of target specs
- identifier-grounding: adversarially fact-checks RM/AM/BASE identifiers in a
draft against the published spec (openehr-assistant MCP or .md twin)
Commands (commands/, disable-model-invocation — user-only, no trigger
competition with the knowledge skills):
- /openehr-specs:amend, :regen-classes, :publish
Each agent/command references its sibling knowledge skill rather than
duplicating it. validate.py now validates agent/command frontmatter; AGENTS,
both READMEs, and CHANGELOG updated; version bumped to 0.2.0 across both
plugin manifests and both marketplace entries.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace absolute local paths in the cross-references guide with links into the openEHR repository (master), with a note to prefer a local sibling checkout when present. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Rounds the
openehr-specsplugin out from skills-only into a full workflow toolkit. Bumps the plugin to 0.2.0 (synced across both plugin manifests and both marketplace entries).Subagents (
agents/)Context-isolated, dispatched by Claude for heavy/multi-file/verification work that would otherwise bloat the main context. Each degrades gracefully when an optional dependency is absent.
spec-reviewerreviewcheck catalog across an entire spec document (master.adoc+ chapters +manifest.json), returns a findings reportspecifications-XX(+AA_GLOBAL)xref-auditor{openehr_*}attribute and<<anchor>>resolves; checks cross-spec deep-link anchors against the target's Markdown twinidentifier-groundingopenehr-assistantMCP or WebFetchCommands (
commands/)User-invoked actions, marked
disable-model-invocation: trueso they never auto-trigger or compete with the knowledge skills./openehr-specs:amend— add an amendment-record entry (version bump, anchors, Jira refs)/openehr-specs:regen-classes— regenerate class tables/diagrams viabmm-publisher(Docker)/openehr-specs:publish— build a local HTML preview via theAA_GLOBALpublisherEach command/agent references its sibling knowledge skill rather than duplicating it.
Design notes
agents//commands/use the default locations (no manifest path declarations — the schema rejects a stringagents, and minimal-manifest is preferred).Validation
scripts/validate.pyextended to validate agent/command frontmatter — passes.claude plugin validatepasses for both the marketplace and the plugin manifest.0.2.0, Agents (3), and the 3 commands discovered.0.2.0entry).🤖 Generated with Claude Code