mcpField
I
insights

insights

github.com/rohitg00Agent workflows2.9k170 installsUpdated 4w ago
Repository

About insights

A skill that analyzes project memory and session history to produce reports on agent work patterns. It is intended for tracking corrections, captured learnings, and productivity trends, with suggestions for recurring issues and stale learnings.

Key features

  1. 01Summarize current session activity
  2. 02Break down learning categories
  3. 03Show correction patterns by category
  4. 04Track productivity across sessions
  5. 05Flag recurring corrections and stale learnings

Use cases

  1. 01Reviewing recent agent session activity
  2. 02Finding categories with frequent corrections
  3. 03Tracking learning capture over time
  4. 04Identifying learnings to turn into rules

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/rohitg00/pro-workflow --skill insights

After installing, type/insightsin Claude Code to confirm it loaded.

Details

From the project README

Session Insights

Surface patterns from learnings and session history.

Trigger

Use when asking "show stats", "how am I doing", "analytics", "insights", "heatmap", or "correction rate".

Data Sources

Gather data from these locations before computing metrics:

# Session history and learnings
cat .claude/LEARNED.md 2>/dev/null || cat CLAUDE.md | grep -A999 "LEARNED"
cat .claude/learning-log.md 2>/dev/null

# Session activity
git log --oneline --since="today" --author="$(git config user.name)"
git diff --stat

A correction is any instance where the user redirected, fixed, or overrode agent output during a session. Count [LEARN] entries and explicit correction markers in session history.

What It Shows

Session Summary
Session Insights
  Duration: 47 min
  Edits: 23 files modified
  Corrections: 2 self-corrections applied
  Learnings: 3 new patterns captured
  Context: 62% used (safe)
Learning Analytics
Learning Insights (42 total)

Top categories:
  Testing     12 learnings (29%)
  Navigation   8 learnings (19%)
  Git          7 learnings (17%)
  Quality      6 learnings (14%)

Most applied:
  #12 [Testing] Run tests before commit — 15 times
  #8  [Navigation] Confirm path for common names — 11 times

Stale learnings (never applied):
  #15 [Editing] Prefer named exports — 0 times (45 days old)
Correction Heatmap
Correction Heatmap

By category (all time):
  ████████████ Testing      34 corrections
  ████████     Navigation   22 corrections
  ██████       Git          18 corrections
  ████         Quality      12 corrections

Hot learnings (most corrected, least learned):
  - [Testing] Mock external deps — corrected 8x, learned 0x
    → Consider: /learn-rule to capture this permanently

Cold learnings (learned but never applied):
  - [Editing] Use named exports — learned 45 days ago, applied 0x
    → Consider removing if no longer relevant
Productivity Metrics
Productivity (last 10 sessions)
  Avg session: 35 min
  Avg edits/session: 18
  Correction rate: 12% (improving)
  Learning capture: 2.1 per session

Guardrails

  • Use actual data from project memory and session history.
  • Surface actionable suggestions, not just numbers.
  • Flag recurring corrections that should become permanent rules.
  • Identify stale learnings that may no longer be relevant.

Output

Formatted analytics report with:

  • Current session stats
  • Category breakdown
  • Most/least applied learnings
  • Correction trends with suggestions
  • Productivity metrics over time

Comments

Markdown supported