explain

A command that explains where locally retained Claude usage went across seven contributor categories, ranked by observed cost and linked to supporting evidence. Locally retained means the data available on the machine, not necessarily the complete billing record.

In plain words
What is it for?
Use it to investigate a subscription week, another date window, exact timestamps, or all supported providers, with optional JSON output.
Why use it?
It helps identify what accounted for recorded spending while clearly stating the limits of the available data.

Command

Install

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.

agentmods
npx agentmods add commands/omrikais/cctally/explain
Clone the repo
git clone --depth 1 https://github.com/omrikais/cctally
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 8,427 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 $0.00000 $0.08427
Opus 5 $0.00000 $0.04214
Sonnet 5 $0.00000 $0.01685
Haiku 4.5 $0.00000 $0.00843

Measured 2d ago against content hash 3954943e9e1c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

explain 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.

docs/commands/explain.md · 331 lines

How it starts

The opening of the file, as written. The whole thing — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cctally explain

Explain where locally retained usage went, over seven contributor classes — four accounting-native and three conversation-derived — ranked by observed cost and backed by evidence you can navigate to.

explain answers one question: which subjects account for the money this window spent. It never gives behavioural advice, never projects a quota percentage onto a slice of spend, and never reports a number without stating the population it was measured over.

Cost coverage: Claude dollar and token totals are transcript-derived lower bounds, not exact /usage billing totals. explain says so on both surfaces: every report states that it explains locally retained cost and tokens.

Usage

cctally explain                                    # the current subscription week, Claude
cctally explain --source all                       # each provider in its own section
cctally explain --window last-week --json          # the stamped JSON envelope
cctally explain --window 2026-08-10..2026-08-16    # an explicit date range
cctally explain --start-at 2026-08-10T00:00:00Z \
                --end-at 2026-08-17T00:00:00Z      # exact instants, half-open
cctally explain --account work --reveal-projects   # one account, project labels shown

Flags

Flag Default Meaning
--source {claude,codex,all} claude Which provider to diagnose. all renders one section per provider, each with its own denominator and verdict.
--account REF none Scope to one account. The ref resolves case-insensitively by label, email or unique key prefix. Rejected with --source all.
--window TOKEN this-week The window to diagnose. Same grammar as cctally diff: this-week, last-week, Nw-ago, this-month, last-month, Nm-ago, last-Nd, prev-Nd, or YYYY-MM-DD..YYYY-MM-DD.
--start-at ISO none Exact window start instant, inclusive. Requires --end-at, and cannot be combined with --window.
--end-at ISO none Exact window end instant, exclusive. Requires --start-at, and cannot be combined with --window.
--speed {auto,standard,fast} auto The Codex pricing tier applied to Codex accounting.
--tz TZ config display.tz Display timezone: local, utc, or an IANA name.
--json off Emit the stamped JSON envelope instead of the terminal report.
--reveal-projects off Show derived project display labels instead of the default project-1, project-2, … aliases.

Read the full file on GitHub · 331 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. 2d ago First seen · 331 lines · 0 tokens per session scan A 3954943e9e1c

Subscribe to this mod's changes

explain is a command published in the GitHub repository omrikais/cctally (5 stars, last pushed 3d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 8,427 tokens. 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.