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/sourikduttanyu/astral/windowgit clone --depth 1 https://github.com/sourikduttanyu/astralWrote 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/sourikduttanyu/astral/window)<a href="https://agentmods.dev/commands/sourikduttanyu/astral/window"><img src="https://agentmods.dev/badge/commands/sourikduttanyu/astral/window.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 | $0.00020 | $0.00397 |
| Opus 5 | $0.00010 | $0.00198 |
| Sonnet 5 | $0.00004 | $0.00079 |
| Haiku 4.5 | $0.00002 | $0.00040 |
Grade A, and why
window 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 3d 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
Astral /astral:window — assert the context window for this project so the warn
bands and the statusline badge match your actual model/tier. Astral can't read
the real window from a hook (the harness/subscription sets it), so this records
your choice.
Argument: $ARGUMENTS — interpret it as:
opusor1mor1000000→ 1000000sonnetor200kor200000→ 200000- a bare integer → that exact token count
auto/reset/ empty → clear the override (Astral falls back to its occupancy-floor auto-detection)
Steps:
- Resolve the argument to an integer (or "clear" for auto/reset/empty). If it's empty and you're acting on an Astral model-change prompt, ask the user which window applies (200K / 1M / keep) via AskUserQuestion first.
- Write that integer to
.astral/windowin the current working directory (create.astral/if needed). For auto/reset, delete.astral/windowif it exists. - Confirm to the user, e.g. "Astral window set to 1,000,000 tokens — bands now fire at 400K / 550K / 700K." (bands are 40/55/70% of the window). For a clear, say auto-detection is restored.
Note: this is per-project (lives in .astral/). ASTRAL_WINDOW env still
overrides it. The change takes effect on your next prompt (when the monitor
re-reads it). Keep it tight.
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.
- 3d ago First seen · 31 lines · 20 tokens per session scan A c339981f9219
window is a command published in the GitHub repository sourikduttanyu/astral (1 stars, last pushed 2mo ago), licensed MIT. It adds 20 tokens to every session and 397 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
smartcompact
Pin important turns before compacting — pins survive compaction via SessionStart hook reinjection.
recall-maintain
Run maintenance on the memory system. Primary function: pressure-based demotion when working memory exceeds the word budget. Secondary: stale entry verification, contradiction checks, open loop review.
recall-init-ids
Add ^tr IDs to untagged memory entries in managed files. This is safe to run multiple times - it only tags entries that don't already have IDs.
recall-init
Initialize the Total Recall memory system in this project.
recall-write
Write a note to memory, applying the write gate. Default destination: daily log.
recall-promote
Review recent daily logs and surface candidates for promotion to registers or CLAUDE.local.md.