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 skills add tyejcoleman/tokenroom --skill skillgit clone --depth 1 https://github.com/tyejcoleman/tokenroomWrote 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/skills/tyejcoleman/tokenroom/skill)<a href="https://agentmods.dev/skills/tyejcoleman/tokenroom/skill"><img src="https://agentmods.dev/badge/skills/tyejcoleman/tokenroom/skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tyejcoleman/tokenroom/skill"><img src="https://agentmods.dev/badge/skills/tyejcoleman/tokenroom/skill.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00074 | $0.04348 |
| Opus 5.5 | $0.00030 | $0.01739 |
| Sonnet 5.5 | $0.00015 | $0.00870 |
| Haiku 4.5 | $0.00007 | $0.00435 |
Grade A, and why
tokenroom 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 2mo 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 — 284 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tokenroom: plan within your budgets
You have two scarce budgets, and a [tokenroom] stamp reporting both arrives with user
prompts when the tokenroom collector is installed:
- Rate-limit windows (account-level):
5h: 34% left, resets 14:00 · 7d: 61% left. Subscription windows are use-it-or-lose-it; work attempted past exhaustion fails (429s) and in-flight work is wasted. - Context headroom (session-level):
ctx: ~38k tokens before compaction. Past the ceiling the conversation is compacted and task continuity degrades.
All percentages in stamps and tools are remaining, never used.
How these budgets actually behave (do not conflate them)
- Quota (the 5h/7d windows) is account-level spend. It refills ONLY at its reset
clock. Low quota → defer heavy work past the reset (
plan_resume). - Context is this session's working memory. No clock refills it — waiting does
nothing. It changes only when compaction fires (automatic, near the ceiling) or the
user runs /clear. Low context → save a
checkpoint, then KEEP WORKING: compaction is automatic and survivable — tokenroom re-injects your checkpoint, repo ground truth, and pins immediately after it. - Never stop to "wait for context to come back at the reset" — that waits for the wrong resource and wastes wall-clock. If a fresh context window would genuinely help the next phase, ask the user to run /compact at a clean boundary; you cannot trigger it yourself.
Before any sizable task
- Estimate its cost. Rough priors: small localized fix ~3k tokens; medium task (docs pass, small feature) ~8–15k; large refactor or multi-file migration ~25k+.
- Call
fit_checkwith that estimate (MCP tool, or trust the stamp if tools are unavailable). Act on the verdict:
| Verdict | Action |
|---|---|
fits |
Proceed normally. |
tight |
Proceed, but cheap-first, batch tool calls, no scope growth, land at a clean boundary. |
exceeds (context) |
Do NOT start as-is: write a checkpoint/handoff first, or split so a piece fits. |
defer (window) |
Do not start. Finish current work at a clean boundary, record a resume plan naming the reset time, and stop. |
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.
- 2mo ago First seen · 284 lines · 74 tokens per session scan A d6cea75e2d57
tokenroom is a skill published in the GitHub repository tyejcoleman/tokenroom (0 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 74 tokens to every session and 4,348 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
tokf-discover
Find missed token savings by scanning AI coding session files for commands that ran without tokf filtering.
calendar
Create, list, and manage calendar events / meetings from the terminal using whatever native CLI is installed (gcalcli for Google Calendar, khal for CalDAV, or AppleScript Calendar on macOS). Use when asked to "schedule a meeting", "create an event", "what's on my calendar", "am I free", "book".
Send and read email from the terminal using whatever native CLI is installed (himalaya, msmtp/mutt/neomutt on Linux, or AppleScript Mail on macOS). Use when asked to "send an email", "email someone", "check my inbox", "read my email", "reply to".
reminders
Create and list reminders / to-dos and time-based nudges using native tooling (AppleScript Reminders on macOS, or ChatCLI's own @scheduler for a timed message back to the user). Use when asked to "remind me", "set a reminder", "add a to-do", "don't let me forget".
send-message
Proactively send a message to the user (or a chat) over a connected messaging platform — Telegram, WhatsApp, Discord, Slack, or a generic webhook. Use when asked to "notify", "ping", "message", "tell", "send", "let me know on Telegram/WhatsApp", or to deliver a result to a chat.
notes
Capture and retrieve quick notes in the OS-native notes app — Apple Notes on macOS, a markdown notes folder on Linux, Sticky Notes/OneNote on Windows. For an Obsidian vault specifically, use the obsidian skill. Use when asked to "make a note", "save this", "jot down", "what did I note about".