chore: publish artifact as part of plan-epic final report

[Walid-peach]

What

Adds an artifact-publishing step to the /plan-epic skill's final report so every epic decomposition also renders a visual dependency-chain breakdown, not just plain text.

Why

Split out of #300 (MON-214) at that PR's own review's request — this doc-only change to .claude/skills/plan-epic/SKILL.md is unrelated to MON-214's scope and belongs in its own PR.

Changes

• .claude/skills/plan-epic/SKILL.md: step 7 now also publishes an HTML Artifact of the sub-issue plan (dependency-ordered cards linking to the created Linear issues), loading artifact-design first and reusing the same scratchpad file path across a planning session so re-publishing after Linear writes updates the same URL.

Testing

Not applicable — this only changes Claude Code skill instructions, no application code.

Risks / Notes

None. No app code, no deploy surface.

Breaking Changes

None.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
mon-elu	Ready	Preview	Aug 3, 2026 9:18pm

[github-actions]

🧪 dbt data-health check — ✅ passed

21:17:04  52 of 62 PASS source_not_null_raw_votes_vote_id ................................ [PASS in 1.36s]
21:17:04  56 of 62 START test source_unique_raw_deputies_deputy_id ....................... [RUN]
21:17:05  53 of 62 PASS source_not_null_raw_votes_voted_at ............................... [PASS in 1.35s]
21:17:05  57 of 62 START test source_unique_raw_votes_vote_id ............................ [RUN]
21:17:05  54 of 62 PASS source_relationships_raw_vote_positions_deputy_id__deputy_id__source_raw_deputies_  [PASS in 1.44s]
21:17:05  58 of 62 START test unique_mart_deputy_scorecard_deputy_id ..................... [RUN]
21:17:05  56 of 62 PASS source_unique_raw_deputies_deputy_id ............................. [PASS in 1.31s]
21:17:05  59 of 62 START test unique_mart_party_alignment_deputy_id ...................... [RUN]
21:17:06  55 of 62 PASS source_relationships_raw_vote_positions_vote_id__vote_id__source_raw_votes_  [PASS in 2.05s]
21:17:06  60 of 62 START test unique_mart_vote_summary_vote_id ........................... [RUN]
21:17:06  57 of 62 PASS source_unique_raw_votes_vote_id .................................. [PASS in 1.32s]
21:17:06  61 of 62 START test unique_stg_deputies_deputy_id .............................. [RUN]
21:17:06  58 of 62 PASS unique_mart_deputy_scorecard_deputy_id ........................... [PASS in 1.31s]
21:17:06  62 of 62 START test unique_stg_votes_vote_id ................................... [RUN]
21:17:06  59 of 62 PASS unique_mart_party_alignment_deputy_id ............................ [PASS in 1.35s]
21:17:07  60 of 62 PASS unique_mart_vote_summary_vote_id ................................. [PASS in 1.37s]
21:17:07  61 of 62 PASS unique_stg_deputies_deputy_id .................................... [PASS in 1.38s]
21:17:07  62 of 62 PASS unique_stg_votes_vote_id ......................................... [PASS in 1.32s]
21:17:09  
21:17:09  Finished running 62 data tests in 0 hours 0 minutes and 28.57 seconds (28.57s).
21:17:09  
21:17:09  Completed successfully
21:17:09  
21:17:09  Done. PASS=62 WARN=0 ERROR=0 SKIP=0 TOTAL=62

Validates prod data health. A failure means current prod data is stale/broken, not necessarily that this PR is wrong.