---
title: "Session Report: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/skills/anthropic-session-report/
description: "Runs a bundled Node analyzer over local Claude Code transcripts and fills an HTML template with token use by project, subagent type and skill, cache breaks and the most expensive prompts, plus a few findings and suggestions."
---

# Session Report: review, setup & limitations · Undominated.ai

> Runs a bundled Node analyzer over local Claude Code transcripts and fills an HTML template with token use by project, subagent type and skill, cache breaks and the most expensive prompts, plus a few findings and suggestions.

[← Explore all skills](/skills/)

OBSERVABILITY / Anthropic

# Session Report

Runs a bundled Node analyzer over local Claude Code transcripts and fills an HTML template with token use by project, subagent type and skill, cache breaks and the most expensive prompts, plus a few findings and suggestions.

 [See setup guidance ↓](#setup)[Original source ↗](https://github.com/anthropics/claude-plugins-official/tree/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report)

SOURCE REVIEW

 Reviewed 2026-10-21
 Evidence 6 linked sources
 Publisher Anthropic
 Licence [Apache-2.0 ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/LICENSE)
 Revision 41f32e407b45
 [Read what was—and wasn’t—checked ↓](#review)

“Generate an explorable HTML report of Claude Code session usage (tokens, cache, subagents, skills, expensive prompts) from ~/.claude/projects transcripts.”

 [Anthropic · upstream description ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/SKILL.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - Its numbers come from a bundled script, analyze-sessions.mjs, which breaks token use down by project, subagent type and skill and lists cache breaks and the top prompts; the model pastes that JSON into the template and adds 3-5 findings, expressed as shares of total tokens, and 1-4 suggestions.
- The analyzer's header documents how it avoids double counting: assistant entries are deduplicated by requestId, and replayed history in resumed sessions by entry uuid.

### A good fit for

 - Find which project, skill or subagent type used the largest share of tokens over the last 7 days, 24 hours, 30 days or all time
- Spot cache-hit rates under 85% and clusters of cache breaks before changing a workflow
- Produce an HTML usage report with drill-downs and optimization notes tied to specific rows

### Weigh up before choosing

 - Last commit touching plugins/session-report/skills/session-report at revision 41f32e40: 2026-04-09 (9dc3809e).
- The SKILL.md calls the report self-contained, but template.html loads the JetBrains Mono font from fonts.googleapis.com and fonts.gstatic.com, so opening the report makes requests to those hosts.
- The embedded JSON includes previews of up to 240 characters of the 100 most expensive prompts, the user messages around them and the transcripts directory path, so a shared report can expose prompt text.
- plugins/session-report has no README, and no file in the repository prints an install line for this plugin at this revision.
- The report is written to the current working directory as session-report-<YYYYMMDD-HHMM>.html, and the intermediate JSON to /tmp/session-report.json.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - plugins/session-report/skills/session-report/SKILL.md
- plugins/session-report/skills/session-report/analyze-sessions.mjs (header, imports, JSON output, prompt previews)
- plugins/session-report/skills/session-report/template.html (head)
- plugins/session-report/LICENSE
- .claude-plugin/marketplace.json (session-report entry)
- README.md

### Our findings

 - session-report declares no allowed-tools: its frontmatter is a name and a one-sentence description.
- analyze-sessions.mjs imports only fs, os, path and readline; it reads ~/.claude/projects by default, or the directory given with --dir.
- The SKILL.md tells the agent to edit the copied template with Edit, not Write, so the template's JS and CSS are kept, and not to open or render the report.
- If the JSON exceeds 2MB, the SKILL.md says to trim top_prompts and cache_breaks to 100 entries before embedding.

### Not established by this review

 - The analyzer was not run and no report was produced or opened.
- analyze-sessions.mjs was read for its header, imports and output functions, not line by line; the body of template.html was not read.
- Not executed: it was not installed or run.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Add a skill to your workflow.

[Upstream setup instructions ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/README.md)
 - No file in the repository prints an install line for session-report; the plugin directory has no README.
- The repository README gives the general form `/plugin install {plugin-name}@claude-plugins-official`, and the marketplace file lists this plugin under the name session-report; no filled-in line is printed, so install.command is null.
- This review installed nothing and analysed no transcripts.

### Before you start

 - Local Claude Code transcripts in ~/.claude/projects
- Node, to run node <skill-dir>/analyze-sessions.mjs

### Compatibility

Claude Code

### Reads local Claude Code transcripts and writes an HTML report into the working directory

 - The session-report SKILL.md lists no allowed-tools; it runs node analyze-sessions.mjs and cp through the session's Bash tool
- Reads ~/.claude/projects/**/*.jsonl, including subagent transcripts and their sibling .meta.json files
- Writes /tmp/session-report.json and ./session-report-<YYYYMMDD-HHMM>.html
- The session's model reads /tmp/session-report.json, with its prompt previews, surrounding user messages, project names, session ids and transcripts path, and writes that JSON into the HTML
- The finished report requests fonts from fonts.googleapis.com and fonts.gstatic.com when opened in a browser

### Cost model

Apache-2.0 per plugins/session-report/LICENSE; the analyzer runs locally and the findings text is written by the session's model

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [session-report SKILL.md ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/SKILL.md) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/SKILL.md Supports: upstreamDescription, summary, whySelected[0], bestFor[0], bestFor[1], bestFor[2], limitations[1], limitations[4], review.findings[0], review.findings[2], review.findings[3], access.permissions[0], access.permissions[2], access.permissions[3], access.cost, install.prerequisites[1], compatibility
- [session-report analyze-sessions.mjs ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/analyze-sessions.mjs) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/analyze-sessions.mjs Supports: whySelected[0], whySelected[1], limitations[2], review.findings[1], access.permissions[1], access.permissions[3], install.prerequisites[0]
- [session-report template.html ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/template.html) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/skills/session-report/template.html Supports: limitations[1], access.permissions[4]
- [session-report LICENSE ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/LICENSE) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/plugins/session-report/LICENSE Supports: license, access.cost
- [Marketplace file, session-report entry ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/.claude-plugin/marketplace.json) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/.claude-plugin/marketplace.json Supports: install.steps[1], summary
- [Repository README, installation ↗](https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/README.md) Checked 2026-10-21 https://raw.githubusercontent.com/anthropics/claude-plugins-official/41f32e407b450055044f167db7ded5e8ff67d203/README.md Supports: install.sourceUrl, install.steps[1], compatibility

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Receipts ↗ Builds a personal Claude Code impact report from this machine's session transcripts and local git history: what shipped, in which projects, and each project's share of usage, as a Markdown report and an HTML receipt in the home directory.](/skills/anthropic-receipts/)[### Arize Instrumentation Health ↗ Read-only audit of exported Arize spans for nine instrumentation defects, including missing token counts behind empty cost dashboards.](/skills/arize-ai-arize-instrumentation-health/)[### Elasticsearch Cluster Health ↗ Triage unassigned Elasticsearch shards through allocation evidence.](/skills/elastic-elasticsearch-cluster-health/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Choose a toolkit](/tools/)
- [Inspect execution hosts](/agents/)
- [Check tool access](/mcp-servers/)
- [Check project dependencies](/check-repo/)
