/status -- Sprint dashboard

The user wants to see the current state of the active sprint.

Arguments

$ARGUMENTS

Instructions

1. Run grainulator.compile to refresh conflicts and warnings.
2. Run grainulator.status to get the current sprint snapshot. Use active_claims and total_claims separately, topic_list, evidence_distribution, type_distribution and the complete conflicts pairs. Read <dir>/compilation.json for per-topic aggregates and warnings; never invent fields from the compact compile response.

3. Display the dashboard:

   Sprint: <slug>
   Phase: <phase> | Claims: <active_claims> active / <total_claims> total, <conflicted> conflicted
   Topics: <list with claim counts>
   
   Evidence tiers:
     stated: <n>  |  web: <n>  |  documented: <n>  |  tested: <n>  |  production: <n>
   
   Claim types:
     constraint: <n>  |  factual: <n>  |  estimate: <n>  |  risk: <n>  |  recommendation: <n>  |  feedback: <n>
   
   Conflicts: <list or "none">
   Warnings: <list or "none">

4. Present next steps from the fresh compilation, following the output contract below.

Verification scope

Carry existing unresolved findings forward, including failures in a pasted review. For each material result, identify the source revision or build, installed package and cache version when relevant, host version, and exact test scope. Mark untested layers explicitly; a source fix does not establish that an installed package or cached plugin is fixed.

Direct MCP success verifies that connection only. Full plugin acceptance must exercise native discovery and the relevant host behavior; subagent acceptance requires an actual nested tool invocation and its result. A subagent's assertion alone is not invocation evidence.

Close a finding only when a check of its original failing path passes on the affected artifact, or when supported evidence explicitly supersedes it. A generic user success report or a ready compiler does not close unrelated failures. Keep remaining gaps visible in status and next actions without inventing work or imposing test quotas.

Host access

Use available grainulator MCP tools, passing the active sprint dir explicitly for evidence operations. If a tool is unavailable, use the local grainulator CLI (or node <checkout>/bin/grainulator.js). Read sibling skill files directly when slash commands are unavailable. Resolve template paths relative to this skill’s checkout when CLAUDE_PLUGIN_ROOT is unset. Optional external connectors are not required for local work; use local code, supplied documents, or available web tools. Do not write managed ledger files directly to bypass a missing MCP connection.

Next-step output

For standalone fetch, setup, or read-only orchestration without an evidence sprint, derive next steps from that task. Do not initialize or compile an unrelated ledger just to produce this footer.

When working in an evidence sprint, use the current compiler's next_actions to present exactly two bullet lists labeled Auto and Manual. Auto is work the agent can continue under existing authorization. Manual is only work requiring the user's decision, access, or action. Classify using the current request and constraints; compiler suggestions never grant permission. Continue authorized Auto work without asking again.

Keep 2–3 useful actions total when available, use short concrete labels and commands where useful, and show None. for an empty group. Do not invent work to fill a quota. Never omit next steps merely because compilation is ready or the answer should be brief. Refresh stale compilation first and exclude work the user removed from scope. When the user asks only for next steps, output only these two lists: no findings recap, counts, reasons, or offer to continue.