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/dumbspacecookie/stickies/btwgit clone --depth 1 https://github.com/dumbspacecookie/stickiesWhat 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 | $0.00016 | $0.00632 |
| Opus 5 | $0.00008 | $0.00316 |
| Sonnet 5 | $0.00003 | $0.00126 |
| Haiku 4.5 | $0.00002 | $0.00063 |
Grade A, and why
btw 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 yesterday.
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
btw
The user invoked /btw with: $ARGUMENTS
This is the frictionless capture path. The user had a thought mid-build and wants it parked so it stops taking up head-space. Your job is to write it down and get out of the way — not to discuss it, act on it, or ask about it.
Run exactly one command:
node --disable-warning=ExperimentalWarning "${CLAUDE_PLUGIN_ROOT}/src/cli.js" add "<text>" -c <category> -i <importance> [-p global]
Text: $ARGUMENTS verbatim, minus any trailing P1/P2/P3/global keyword the user
appended as a directive. Do not rewrite, summarise, or "improve" their phrasing — a note
they don't recognise later is a note they won't trust. Fix nothing but obvious typos.
Category — infer from the text, defaulting to todo:
todo— a task, a fix, something to come back to. This is the default, because/btwis overwhelmingly used to park work.blocker— they are stuck on something external (missing cred, failing CI, waiting on a person).decision— they chose X over Y and the reasoning should survive.context— background worth keeping that isn't actionable.preference— a stable statement about how they want things done.
Importance — default P1. This differs from /stickies add on purpose: a /btw note is
by definition something the user wants to see again, and only P1 renders in full in the
next session's digest and in the statusline (P2 is truncated, P3 is a bare count). Downgrade
to P2/P3 only if the user explicitly says so, or if the note is plainly non-actionable
background (context/preference → P2).
Scope — project (the cwd) unless the user says global, in which case pass -p global.
After it's written, confirm in one short line — what was parked, its priority, and the id, e.g.:
Parked (P1 todo) · "auth is probably leaking on the public route" ·
a1b2c3d4-…
Then stop. Do not expand on the note, do not offer to start working on it, do not ask follow-up questions. If the user wanted to act on it now, they wouldn't have said "btw". The whole value of this command is that it costs the user nothing to use mid-flow.
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.
- yesterday First seen · 50 lines · 16 tokens per session scan A 3cf8361481a6
btw is a command published in the GitHub repository dumbspacecookie/stickies (1 stars, last pushed 29d ago), licensed MIT. It adds 16 tokens to every session and 632 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
everme-help
Print a concise EverMe plugin status + reference card.
flow-nexus-neural
Train and deploy neural networks in distributed sandboxes.
flow-nexus-auth
Flow Nexus authentication and user management.
memcord-save-progress
Auto-summarize and save conversation progress.
validate-prd
Validate an existing PRD against BMAD standards - comprehensive review for completeness, clarity, and quality.
editorial-review-structure
Structural editor that proposes cuts, reorganization, and simplification while preserving comprehension.