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 commands/vsladkov/claudex-stereo/plan-stategit clone --depth 1 https://github.com/vsladkov/claudex-stereoWrote 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.
[](https://agentmods.dev/commands/vsladkov/claudex-stereo/plan-state)<a href="https://agentmods.dev/commands/vsladkov/claudex-stereo/plan-state"><img src="https://agentmods.dev/badge/commands/vsladkov/claudex-stereo/plan-state.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00023 | $0.00933 |
| Opus 5 | $0.00012 | $0.00466 |
| Sonnet 5 | $0.00005 | $0.00187 |
| Haiku 4.5 | $0.00002 | $0.00093 |
Grade A, and why
plan-state scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
!node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --metadata
Raw slash-command arguments:
$ARGUMENTS
The preamble shows the default slot's metadata only, so a flagged invocation never pays for a
plan document it will not use. With no arguments, run
node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state and present that full output
(metadata plus the stored plan document) instead of the preamble. With --list, --compare, or
an explicit --slot,
run the requested companion command and present that output instead.
Apply at most one action:
- With
--list, runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --listand present the slot inventory verbatim.--listdoes not combine with--slot. - With
--compare <slotA> <slotB>, runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --compare <slotA> <slotB>and present the comparison verbatim: both metadata blocks and the plan diff.--comparenames both slots, so it does not combine with--slotor another action. Both slots must hold a stored plan; on a missing-slot error, relay the message and point the user at/stereo:plan-state --list. When the diff is suppressed, relay the export hint instead of attempting a diff yourself. - With
--open, runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --open [--slot <slot>]and relay the exported path and whether VS Code opened. - With
--clearand an explicit slot, first runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --metadata --json --slot <slot>and use its summary, verdict, round, andupdatedAtto describe that exact slot in the confirmation. Then useAskUserQuestionexactly once withDelete the plan in slot <slot>andKeep it (Recommended). Without an explicit slot, retain the existing question:Delete the stored plan and any implementation recordandKeep it (Recommended). Runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --clear [--slot <slot>]only when deletion is confirmed. - With
--mark-implemented, runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --mark-implemented [--slot <slot>]and note that this marker is normally set automatically by/stereo:implementafter a successful full phase. - With an explicit
--slotand no action, runnode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" plan-state --slot <slot>.
Present the full command output to the user. Do not summarize or condense it.
When the output includes the stored plan document (the no-argument and --slot reads), render
it as Markdown while preserving its contents verbatim; action outputs (--list, --clear,
--mark-implemented, --metadata) contain no plan document — do not fetch one to show.
Clearing a slot removes the implementation record only when that record belongs to the cleared
slot. When --clear also removes an implementation record, preserve the extra status and path
lines. Relay the Kept the implementation record... line verbatim when it is present.
Relay implementedAt when present; it means a full implementation phase completed with an
accepted review, not that the work was committed or merged.
If no plan is stored, relay the message and point the user to /stereo:plan.
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.
- 4d ago First seen · 60 lines · 23 tokens per session scan A 3c5642e75ad9
plan-state is a command published in the GitHub repository vsladkov/claudex-stereo (3 stars, last pushed 26d ago), licensed Apache-2.0. It adds 23 tokens to every session and 933 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.
Other commands, from other repositories
rest
Maude's end-of-session ritual — digest + tomorrow's starting point + close-the-loop check + full save fan-out. Run before you stop, so the next session opens cleanly. Session-end is the right time to pay network cost, so Tier 2 writes happen here when registered.
save
Save the current session digest. Fans out across every writable memory tier per the house-map. Tier 0 always; Tier 1 if reachable; Tier 2 only if registered writable with auth. Each tier independent — failures degrade gracefully.
wake
Maude's session-start ritual — brief + house-walk + 1-3 things to know first. Runs at session-start manually if SessionStart hook didn't catch you, or anytime you need a clean re-orient. Cheap by design — Tier 0 always, Tier 1 if cached-up, Tier 2 never.
notice
Maude surfaces patterns from the turn-by-turn trace — recurring topics, repeated mistakes, time-of-day patterns, sessions that keep ending in the same place.
teach
Tell Maude something about yourself directly — how you work, your rhythm, what help you want — and she records it in her cross-project profile of you. Use when you want her to KNOW a fact rather than wait for her to infer it. Tier 0, user-global, no network.
verify
Run Maude's programmatic project audit — version consistency, JSON validity, link integrity, header dates, watch-list paths, worn-framing scan. Catches "ready" before it's actually ready.