| name | sprang-team |
| description | Browse and write team annotations attached to graph nodes — institutional knowledge, decision history, ownership. Use when the user says "/sprang-team", "add annotation", "team knowledge", "who owns this", or "document this decision". |
Follow the detailed instructions in .windsurf/workflows/sprang-team.md.
Quick steps:
- Check
.sprang/knowledge-graph.json exists — if not, tell the user to run /sprang-analyze first.
- List all files in
.sprang/annotations/ to find every annotated node.
- For each annotation file, read YAML frontmatter:
node_id, node_label, annotated_at, tags.
- If no annotations exist: suggest candidates using step 8 below, then offer to write the first one.
- Present annotation index table sorted by
annotated_at descending.
- If
$ARGUMENTS is a node path or ID: call sprang_why + sprang_node for that node and display annotation, decision context, and current structural state side-by-side.
- Staleness detection — for each annotated node:
- Call
sprang_node to check in_degree/out_degree and node.decision_context.last_changed
- Use
last_changed if present; otherwise fall back to graph.stats.generated_at as the "last analyzed" proxy
- If the fallback date >
annotated_at: flag as ⚠️ possibly stale (graph was re-analyzed since annotation was written)
- Report a staleness summary table
- Suggest unannotated candidates: nodes with
risk_score > 0.6 OR in_degree > 10 that have no annotation file.
- Offer to write: use
sprang_annotate MCP tool to create .sprang/annotations/<node-id>.md with purpose, decisions, caveats, and ownership.
- Optionally save annotation report to
docs/ANNOTATIONS.md.
