Shaktra Status Dash

/shaktra:status-dash — Project Dashboard

When this skill is invoked, immediately execute all sections below in order and present the results as a formatted dashboard. Do not ask clarifying questions — just run the checks and display the output.

Read-only — never create, modify, or delete any file. Use ${CLAUDE_PLUGIN_ROOT} to locate the installed plugin directory.

Section 1 — Plugin Info + Version Check

Run the version check script via Bash:

python3 ${CLAUDE_PLUGIN_ROOT}/scripts/check_version.py ${CLAUDE_PLUGIN_ROOT}

The script outputs a JSON object. Parse the status field and display accordingly:

If the JSON contains "error", display: "Plugin not configured — plugin.json not found."

Section 2 — Sprint Health

1. Read .shaktra/sprints.yml.
2. Find the sprint where status is active (or the most recent sprint).
3. Display:
   • Sprint ID and goal
   • Start date → end date
   • Days remaining (compute from end date vs today)
   • Capacity: committed points vs capacity points
4. If velocity_history exists, show the last 3 entries as a trend line (e.g., Velocity trend: 18 → 21 → 24).
5. If no sprints file or no active sprint: display "No active sprint — run /shaktra:tpm to plan one."

Section 3 — Story Pipeline

1. Glob .shaktra/stories/*/handoff.yml — read each file.
2. Extract story_id and current_phase from each handoff.
3. Group stories by phase in pipeline order: pending → plan → tests → code → quality → complete → failed.
4. Display as a table:

| Phase    | Count | Stories          |
|----------|-------|------------------|
| pending  | 2     | ST-004, ST-005   |
| plan     | 1     | ST-003           |
| code     | 1     | ST-002           |
| complete | 1     | ST-001           |

5. Only show phases that have at least one story.
6. If a story has blocked: true or blocker field set, append "(blocked)" after its ID.
7. If no stories exist: display "No stories found — run /shaktra:tpm to create stories."

Section 4 — Quality Summary

1. Read all .shaktra/stories/*/handoff.yml files.
2. For each, check for quality_findings (a list of findings with severity and resolved fields).
3. Count unresolved findings (where resolved is false or absent) grouped by severity: P0, P1, P2, P3.
4. Read .shaktra/settings.yml → quality.p1_threshold for the P1 limit.
5. Determine merge-blocking status:
   • Blocked if any unresolved P0 exists
   • Blocked if unresolved P1 count exceeds p1_threshold
   • Clear otherwise
6. Display:

| Severity | Unresolved |
|----------|------------|
| P0       | 0          |
| P1       | 1          |
| P2       | 3          |
| P3       | 2          |

Merge status: Clear (P1: 1/2 threshold)

7. If no quality findings exist across any story: display "No quality findings recorded."

Section 5 — Memory Health

1. Read .shaktra/memory/principles.yml → count entries by confidence band. If file is missing or empty, count is 0.
   • High: confidence >= 0.7
   • Moderate: confidence >= 0.4 and < 0.7
   • Low: confidence >= 0.2 and < 0.4
2. Read .shaktra/memory/anti-patterns.yml → count active entries. If missing or empty, count is 0.
3. Read .shaktra/memory/procedures.yml → count active entries. If missing or empty, count is 0.
4. Read .shaktra/settings.yml → memory.max_principles, memory.max_anti_patterns, memory.max_procedures for limits.
5. Display:
   • Principles: {total_count} ({high}↑ {moderate}~ {low}↓)
   • Anti-patterns: {count}
   • Procedures: {count}
6. If any file is at 80% or more of its max limit, add warning: "Approaching limit — lowest confidence entries will be archived."

Section 6 — Analysis Progress

1. Check if .shaktra/analysis/manifest.yml exists. If not, skip this section entirely.
2. Read the manifest — extract each dimension entry (D1 through D9) with its status.
3. Display:

| Dimension | Name            | Status    |
|-----------|-----------------|-----------|
| D1        | Architecture    | complete  |
| D2        | Practices       | complete  |
| D3        | Dependencies    | pending   |
| ...       | ...             | ...       |

4. Show summary: Completed: {n}/9 dimensions

Graceful Degradation

Each section degrades independently:

• If a file is missing or unreadable, show a short message for that section and continue to the next.
• If .shaktra/ directory does not exist, show Section 1 (Plugin Info + Version Check) only, then display: "Project not initialized — run /shaktra:init to set up."
• Never fail the entire dashboard because one section has missing data.

Output Format

Present the dashboard as a structured report:

## Shaktra Dashboard

### Plugin Info
Version: 0.1.0
Status: Up to date (v0.1.0)

### Sprint Health
Sprint: S-003 — "Auth module"
Dates: 2025-01-06 → 2025-01-20 (5 days remaining)
Capacity: 21/30 points committed
Velocity trend: 18 → 21 → 24

### Story Pipeline
| Phase    | Count | Stories        |
|----------|-------|----------------|
| code     | 1     | ST-002         |
| complete | 2     | ST-001, ST-003 |

### Quality Summary
| Severity | Unresolved |
|----------|------------|
| P0       | 0          |
| P1       | 1          |
| P2       | 3          |
| P3       | 2          |
Merge status: Clear (P1: 1/2 threshold)

### Memory Health
Principles: 12 (8↑ 3~ 1↓)
Anti-patterns: 2
Procedures: 1

### Analysis Progress
Not applicable — no analysis manifest found.