Analyze Session Transcript

Analyze a Figma MCP test session transcript and produce a structured efficiency/error audit. After large Figma sessions (50+ tool calls), run this skill to capture learnings and track improvement over time.

Phase 1: Locate and Ingest Transcript

Session manifest

A manifest at .claude/analysis/sessions.json tracks all sessions and their analysis status:

{
  "sessions": {
    "<session-id>": {
      "sessionType": "figma" | "dev" | "empty",
      "skip": true,              // present on dev/empty sessions
      "toolCalls": 56,
      "figmaToolCalls": 48,
      "durationMinutes": 20,
      "sourceModified": 1710000000.00,  // mtime of source JSON
      "analysis": "figma-mcp-session4-analysis.md",  // only if analyzed
      "analyzedAt": 1710000000.00       // mtime of analysis file
    }
  }
}

Sessions with sessionType: "figma" (at least 1 mcp__Figmagent__* tool call) are candidates for analysis. Sessions with sessionType: "dev" or "empty" are skipped.

Picking the session to analyze

1. First, ensure all sessions are extracted: Run bun extract-sessions --compact --no-thinking to extract any new/updated sessions (mtime-based skipping is built in). For sessions from other projects, use --file <path> to point at an external JSONL file directly (e.g. bun extract-sessions --file ~/.claude/projects/-Users-foo-Github-other-project/<session-id>.jsonl --compact --no-thinking --include-agents).

2. Then, refresh the manifest: Run the manifest update script (see below) to discover new sessions and check for stale analyses.

3. Pick the target session:

   • If a file path argument was provided, use that specific session.
   • Otherwise, read .claude/analysis/sessions.json and find Figma sessions that need analysis:
     • sessionType: "figma" AND no analysis field → new, needs analysis
     • sessionType: "figma" AND sourceModified > analyzedAt → updated, needs re-analysis
   • Pick the oldest unanalyzed session first (analyze in chronological order).
   • If all Figma sessions are analyzed and up-to-date, report "All sessions analyzed" and stop.

4. Analyze one session at a time to keep context manageable. After completing one analysis, the user can run the skill again to analyze the next.

Manifest update script

Run this Python snippet via Bash to refresh the manifest before analysis:

python3 -c "
import json, os, glob

sessions_dir = '.claude/sessions-json'
analysis_dir = '.claude/analysis'
manifest_path = f'{analysis_dir}/sessions.json'

# Load existing manifest or start fresh
try:
    with open(manifest_path) as fh:
        manifest = json.load(fh)
except (FileNotFoundError, json.JSONDecodeError):
    manifest = {'sessions': {}}

# Scan all session JSONs
for f in sorted(glob.glob(f'{sessions_dir}/*.json')):
    with open(f) as fh:
        data = json.load(fh)
    sid = data['sessionId']
    m = data['metadata']
    tools = m.get('uniqueTools', [])
    figma_tools = [t for t in tools if 'Figmagent' in t]
    tc = m['toolCallCount']
    source_mtime = round(os.path.getmtime(f), 2)

    # Preserve existing analysis mapping if present
    existing = manifest['sessions'].get(sid, {})

    entry = {
        'toolCalls': tc,
        'figmaToolCalls': len(figma_tools),
        'durationMinutes': round(m['duration']['minutes']),
        'sourceModified': source_mtime,
    }

    if tc == 0:
        entry['sessionType'] = 'empty'
        entry['skip'] = True
    elif len(figma_tools) > 0:
        entry['sessionType'] = 'figma'
        if 'analysis' in existing:
            entry['analysis'] = existing['analysis']
            # Check if analysis file still exists and get its mtime
            af = f'{analysis_dir}/{existing[\"analysis\"]}'
            if os.path.exists(af):
                entry['analyzedAt'] = round(os.path.getmtime(af), 2)
    else:
        entry['sessionType'] = 'dev'
        entry['skip'] = True

    manifest['sessions'][sid] = entry

with open(manifest_path, 'w') as fh:
    json.dump(manifest, fh, indent=2)

# Report
figma = {k:v for k,v in manifest['sessions'].items() if v.get('sessionType') == 'figma'}
needs = {k:v for k,v in figma.items() if 'analysis' not in v or v.get('sourceModified',0) > v.get('analyzedAt',0)}
print(f'Figma sessions: {len(figma)}, needs analysis: {len(needs)}')
for sid, v in sorted(needs.items(), key=lambda x: x[1]['sourceModified']):
    status = 'new' if 'analysis' not in v else 'updated'
    print(f'  {sid}  {v[\"toolCalls\"]:>4} calls  {v[\"figmaToolCalls\"]:>2} figma  ({status})')
"

After completing analysis

Update the manifest entry for the analyzed session:

• Set analysis to the filename (e.g. figma-mcp-session10-analysis.md)
• Set analyzedAt to the current time

This can be done by reading the manifest, updating the entry, and writing it back.

5. If no extracted JSON exists yet, run bun extract-sessions --compact --no-thinking to extract all sessions from the Claude Code session store. This produces structured JSON files in .claude/sessions-json/. Use --file <path> for sessions from other projects.

6. Reading the JSON transcript (produced by scripts/extract-sessions.ts):

   • Read the file. If >500 lines, read in 500-line chunks.
   • The format is an ExtractedSession object with this structure:
   jsonCopy

   {
     "sessionId": "uuid",
     "extractedAt": "ISO-8601",
     "metadata": {
       "cwd": "/path/to/project",
       "branch": "branch-name",
       "version": "claude-code-version",
       "messageCount": 120,
       "toolCallCount": 89,
       "uniqueTools": ["create", "apply", "get", ...],
       "duration": { "start": "ISO-8601", "end": "ISO-8601", "minutes": 80 }
     },
     "messages": [
       {
         "role": "user" | "assistant" | "system",
         "timestamp": "ISO-8601",
         "content": [
           { "type": "text", "text": "..." },
           { "type": "tool_use", "id": "toolu_xxx", "name": "create", "input": { ... } },
           { "type": "tool_result", "tool_use_id": "toolu_xxx", "content": "...", "is_error": true }
         ],
         "model": "claude-opus-4-6",
         "usage": { "input_tokens": 1234, "output_tokens": 567 },
         "uuid": "msg-uuid",
         "parentUuid": "parent-msg-uuid"
       }
     ],
     "subAgents": {
       "agent-uuid": { /* same ExtractedSession structure */ }
     }
   }
   

   Key fields for analysis:

   • metadata.toolCallCount and metadata.uniqueTools — pre-computed totals
   • metadata.duration.minutes — session length
   • Content blocks with type: "tool_use" — tool calls (.name = tool name, .input = params)
   • Content blocks with type: "tool_result" — results (.is_error = true for failures, .content = error message or result)
   • subAgents — nested sub-agent sessions (same structure, analyze separately then merge)
   • usage on assistant messages — token consumption per turn

7. Three-pass approach (critical for large transcripts — 800+ events):

   • Pass 1 (Extract): Read in chunks. For each message, scan content blocks. For each tool_use block, record: timestamp, tool name, input params (extract nodeId if present). For each tool_result block, record: tool_use_id, is_error, error message snippet. Output a compact one-line-per-tool-call summary. This reduces 300KB → ~15KB.
   • Pass 2 (Analyze): Over the compact summary, compute all metrics and identify patterns.
   • Pass 3 (Detail): For each flagged issue/error pattern, go back to the original transcript to extract specific context (full error messages, parameter values, cascading effects).

8. For HTML transcripts (fallback if no JSON available and extract-sessions cannot run):

   • Read page by page (each HTML file is one page).
   • Extract tool call blocks using pattern matching: look for tool names, parameters, results, and error messages.

Phase 2: Compute Metrics

Calculate these standard metrics from the extracted events:

Session Overview

• Duration: end time - start time
• Total events: count of all events
• Total tool calls: use metadata.toolCallCount or count tool_use content blocks
• Total errors: count tool_result blocks where is_error: true
• Reconnections: count tool_use blocks where name is join_channel (subtract 1 for initial join)
• Context overflows: detect by looking for continuation summaries or session restart markers
• Phases completed: identify distinct work phases from the transcript

Tool Call Distribution Table

For each unique tool name:

• Count total invocations
• Note patterns:
  • "no batch version" if >20 sequential calls to same tool
  • "N redundant re-inspections" if same node ID appears in multiple get calls
  • "N failed" if error count > 0

Error Extraction

• Group errors by error message pattern (normalize variable parts like node IDs)
• Count cascading errors: when one error in a parallel batch causes all parallel calls to fail, count the root error separately from cascaded ones
• Identify root cause vs symptom errors

Efficiency Signals — Detect These Patterns

1. Sequential same-tool runs: 5+ consecutive calls to the same tool → batch candidate. Record: tool name, run length, what a batch version would look like.

2. Inspect-after-create: create or clone_node immediately followed by get on the created node → indicates create response should be richer. Count occurrences.

3. Delete-recreate cycles: delete_node/delete_multiple_nodes followed by create for the same purpose → indicates missing modify capability or wrong initial approach.

4. ToolSearch overhead: total ToolSearch calls, percentage of all calls, failed searches (found wrong tools or 0 results).

5. Redundant re-inspections: same node ID appearing in multiple get calls → count unique nodes vs total get calls.

6. Timeout cascades: 3+ consecutive timeouts → connection loss not detected fast enough.

7. Error retry storms: same error repeated 3+ times → fail-fast rule violated.

Phase 3: Cross-Session Comparison

1. Read the improvement tracker at .claude/analysis/improvement-tracker.md
2. Read the most recent previous analysis from .claude/analysis/ (by filename number)
3. Compute deltas:
   • Waste percentage change
   • Error rate change
   • ToolSearch overhead change
   • New tools used that didn't exist in previous session
   • Recurring issues vs new issues
4. Check which previously-identified issues were addressed:
   • Tool exists now that was flagged as missing? → Mark as implemented
   • Error pattern from previous session not observed? → Mark as verified
   • Same issue still present? → Increment sessions affected count

Phase 4: Generate Analysis Document

Write the analysis to .claude/analysis/figma-mcp-session<N>-analysis.md where N is auto-incremented based on existing files in the directory.

Use this exact template structure (matching the format of existing session 1 and session 2 analyses):

# Figma MCP Session <N> Analysis

## Session Overview

- **Transcript**: `<filename>`
- **Duration**: <duration>
- **Total tool calls**: <count>
- **Total errors**: <count>
- **Reconnections**: <count>
- **Context restarts**: <count>
- **Task**: <brief description>

## Metrics

| Metric | Previous Session | This Session | Change |
|---|---|---|---|
| Total Figma tool calls | ... | ... | ... |
| Meta/overhead calls | ... | ... | ... |
| ToolSearch calls | ... | ... | ... |
| Estimated waste % | ... | ... | ... |

## Tool Call Distribution

| Tool | Calls | Notes |
|---|---|---|
| ... | ... | ... |

## Efficiency Issues

### 1. <Issue title> (saves ~N calls)

<Description of the pattern observed. Include specific numbers — how many consecutive calls, which nodes, what the agent was trying to do.>

**Pattern observed:** <concrete example from the transcript>

**Root cause:** <why this happened — missing tool, wrong default, agent behavior>

**Proposed fix:** <specific actionable recommendation>

**Estimated savings:** ~N calls → ~M calls.

### 2. ...

## Error Analysis

### 1. <Error category> (<N> failures, ~<M> minutes lost)

<Description. Include the exact error message. Trace cascading effects.>

**Agent recovery:** <how the agent responded — did it fail fast? retry too many times?>

**Fix needed:** <specific code or behavior change>

### 2. ...

## What Worked Well

1. **<Tool/pattern>.** <Why it was effective, with specific numbers.>
2. ...

## Priority Improvements

### Tool Changes (ranked by call savings)

1. **<tool name>** — <what it should do>. Saves ~N calls per session.
2. ...

### Agent Skill Updates

1. **<behavior change>** — <description>.
2. ...

Phase 5: Update Improvement Tracker

Update .claude/analysis/improvement-tracker.md:

1. Add new issues: For each efficiency issue or error pattern identified in this analysis that doesn't already exist in the tracker:

   • Assign an ID: [CATEGORY-NNN] where CATEGORY is TOOL, BUG, AGENT, or INFRA
   • Auto-increment NNN within the category
   • Set status to identified
   • Set priority based on estimated call savings: P0 (>50 calls), P1 (10-50 calls), P2 (<10 calls)
   • Classify as auto-fixable if it matches a known fix pattern (see Phase 6)

2. Update existing issues: For each tracker entry:

   • If the issue was not observed in this session and the fix is confirmed working → advance to verified, move to Resolved Issues
   • If the issue recurred → add this session number to "Sessions affected"
   • If a tool was implemented that addresses the issue → advance to implemented

3. Deduplication: Match new findings against existing entries by:

   • Category match
   • Tool name match (if issue references a specific tool)
   • Key phrase match (substring: "batch", "async", "timeout", "coercion", etc.)
   • If match found → increment occurrence count, don't create duplicate

4. Update Metrics Over Time table: Add a row for this session.

5. Update "Last updated" date and "Sessions analyzed" count.

6. Update the session manifest (.claude/analysis/sessions.json): Set the analysis field to the analysis filename and analyzedAt to the current time for the session just analyzed. This marks it as complete so the next /analyze-session invocation skips it.

Phase 6: Generate Fix Plans (if applicable)

For issues marked auto-fixable: yes in the tracker, generate implementation plans. Plans go to .claude/plans/<date>-<issue-id>.md.

Safe Fix Patterns (allowlist)

Only generate plans for these well-understood patterns:

sync-to-async

• Trigger: Error message contains "Cannot call with documentAccess: dynamic-page" or "Use node.setXxxAsync instead"
• Fix: Find the sync call in plugin source, replace with async equivalent
• Plan content: Exact file path, line number, old code → new code
• Example: node.textStyleId = id → await node.setTextStyleIdAsync(id)

type-coercion

• Trigger: Error message contains "expected number, received string" or similar type mismatch
• Fix: Add toNumber() coercion in the plugin handler (helper already exists in src/figma_plugin/src/helpers.js) or add .or(z.string().transform(Number)) to the Zod schema in the MCP tool handler
• Plan content: File path, parameter name, Zod schema change or toNumber() wrapping

missing-batch-tool

• Trigger: Single-item tool called 20+ times consecutively
• Fix: Create batch variant following existing patterns (set_multiple_text_contents, delete_multiple_nodes)
• Plan content: Tool specification (name, parameters, behavior) for use with /add-mcp-tool skill. Include the proposed JSON input format based on observed usage patterns.

Plan Format

# Fix: [ISSUE-ID] <title>

**Pattern**: <sync-to-async | type-coercion | missing-batch-tool>
**Priority**: <P0 | P1 | P2>
**Estimated savings**: <N calls/session>

## Changes

### File: `<path>`
- Line N: `<old code>` → `<new code>`

## Verification
- [ ] Run `bun run lint`
- [ ] Run `bun run test`
- [ ] Run `bun run build:plugin`
- [ ] Test in a Figma session

Important: The skill NEVER applies code changes directly. It only generates plan files and marks issues as planned in the tracker. The user reviews and triggers implementation.

Notes

• If the transcript is too large to fit in context even with the 3-pass approach, focus on the tool call distribution and error extraction (Phases 2a-2b) and skip detailed efficiency pattern analysis for the middle sections.
• Always validate numbers: total tool calls should equal sum of distribution table. Error count should match error analysis section.
• When comparing sessions, normalize for scope differences (session 2 had 26% more tool calls because the task was larger, not because it was less efficient).
• The analysis document is committed to git — it serves as a permanent record of the session and its learnings.