claude-code-autoconfig: Command for Claude Code

.claude/commands/analyze-session.md

analyze-session is a command for Claude Code from design-and-deliver/claude-code-autoconfig. It costs 13 tokens per session (958 once invoked), scanned A, original, MIT.

A session review that summarizes where the session's text-generation budget was spent and gives an efficiency assessment. The underlying summary is produced deterministically before the review is shown.

In plain words
What is it for?
Use it with a session ID prefix, a project/session label from the usage report, or a transcript path. It can also analyze the current session when no argument is supplied.
Why use it?
It helps explain why a session used its budget and identify changes that could make future sessions more efficient. It avoids repeating the analysis when a prepared summary is already available.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: reads .claude/ paths; mentions subagents.

This is design-and-deliver/claude-code-autoconfig's own configuration. It tells Claude Code how to work on claude-code-autoconfig itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-code-autoconfig configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is Bash(node .claude/hooks/token-guard.js --analyze:*).

Reuse

Borrowing it

Nothing to install: this file belongs to design-and-deliver/claude-code-autoconfig. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/design-and-deliver/claude-code-autoconfig/main/.claude/commands/analyze-session.md
Clone the repo
git clone --depth 1 https://github.com/design-and-deliver/claude-code-autoconfig

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for analyze-session

README.md
[![agentmods](https://agentmods.dev/badge/commands/design-and-deliver/claude-code-autoconfig/analyze-session.svg)](https://agentmods.dev/commands/design-and-deliver/claude-code-autoconfig/analyze-session)
Your own site
<a href="https://agentmods.dev/commands/design-and-deliver/claude-code-autoconfig/analyze-session"><img src="https://agentmods.dev/badge/commands/design-and-deliver/claude-code-autoconfig/analyze-session.svg" alt="Measured on agentmods" height="20"></a>
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 958 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00013 $0.00958
Opus 5 $0.00006 $0.00479
Sonnet 5 $0.00003 $0.00192
Haiku 4.5 $0.00001 $0.00096

Measured 6d ago against content hash 60eaa2c963c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

analyze-session 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 6d 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.

.claude/commands/analyze-session.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Analyze Session

Explain where a session's tokens went and what to change next time. The heavy lifting is deterministic (the --analyze digest — pure node, no LLM, ~free) and has ALREADY RUN: the injected output below was generated at prompt time, before the model saw this prompt. The model interprets ONLY the digest.

Argument: $ARGUMENTS — a session-id prefix, the exact project/sid8 label shown on /usage-report's LAST 5 HOURS rows (both forms resolve), or a full transcript path.

Digest (injected at prompt time — zero model round trips)

!node .claude/hooks/token-guard.js --analyze $ARGUMENTS

Step 1: check the injected digest

The digest above already ran — re-running it via Bash wastes the round trip the injection exists to save. Route on its first line:

  • Starts with SESSION → healthy; go straight to Step 2.
  • Says usage: → the command was invoked with no argument: derive the CURRENT session's transcript path (~/.claude/projects/<project-slug>/<session-id>.jsonl, same derivation as /usage-report) and run node .claude/hooks/token-guard.js --analyze <path> via Bash — the only case that needs a model-driven run.
  • Says ambiguous → show the candidate list and ask which one was meant.
  • Says no transcript matches → report that verbatim and point at /usage-report's row labels as the source of valid sids.

Step 2: interpret — the digest ONLY

Hard rule: never read the raw transcript. Analyzing a 40M-token session must not itself load 40M tokens — if the digest can't answer something, say what's missing instead of digging. (Targeted transcript digging is future work, behind an explicit user approval.)

Read the full file on GitHub · 72 lines

Changes

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.

  1. 6d ago First seen · 72 lines · 13 tokens per session scan A 60eaa2c963c8

Subscribe to this mod's changes

analyze-session is a command published in the GitHub repository design-and-deliver/claude-code-autoconfig (2 stars, last pushed 2d ago), licensed MIT. It adds 13 tokens to every session and 958 once invoked, about $0.0001 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.