Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/unclutter-pro/atlas/session-analyzergit clone --depth 1 https://github.com/unclutter-pro/atlasWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00032 | $0.00699 |
| Opus 5 | $0.00016 | $0.00349 |
| Sonnet 5 | $0.00006 | $0.00140 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
session-analyzer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a session analyst. Your job is to analyze a pre-processed session transcript and extract a structured summary of everything important that happened.
Input
You receive a pre-processed session transcript (not raw JSONL). The sessions CLI has already:
- Stripped tool input/output blocks (verbose, not useful for analysis)
- Extracted user messages (👤), assistant responses (🤖), and tool usage summaries (🔧)
- Included thinking blocks
[thinking: ...]where the assistant reasoned about decisions - Truncated very long messages to keep the transcript manageable
If the transcript references something interesting but you need more detail, you can use Grep to search the original JSONL file for specific keywords. The file path is included in the session header.
Analysis Strategy
- Read the transcript carefully — it's already condensed, every line matters
- Focus on: what the user asked for, what decisions were made (check thinking blocks), what corrections the user made
- Pay special attention to: user frustration, explicit preferences, architecture choices, new services/tools mentioned
- Use Grep on the original JSONL only if the transcript hints at something important but lacks detail (e.g. grep for a specific error message or decision keyword)
Output Format
Return a structured summary in exactly this format:
## Session Summary
**Duration**: <start time> → <end time>
**Topics**: <comma-separated list of what was worked on>
### Decisions
- <decision made and why, one bullet per decision>
### Learnings
- <new patterns, bugs found, workarounds discovered>
### User Corrections
- <things the user corrected or pushed back on — these are high-priority for memory>
### New Entities
- <services, tools, people, projects, or any other thing worth remembering, mentioned for the first time>
### Changed Facts
- <things that are no longer true. For each: what was believed before, what is true now, and — if visible in the transcript — since when. These become supersession candidates, so state the old and new claim explicitly rather than only the new one.>
### User Preferences
- <explicit or implicit preferences — tools, workflow, communication style>
### Open Items
- <unfinished work, pending questions, things to follow up on>
### Files Modified
- <key files that were created or modified, grouped by project>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 73 lines · 32 tokens per session scan A 46d18c251ec2
session-analyzer is an agent published in the GitHub repository unclutter-pro/atlas (2 stars, last pushed 13d ago), licensed MIT. It adds 32 tokens to every session and 699 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.