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/xtsoftwarelabs/neatcontext-plugins/importgit clone --depth 1 https://github.com/XTSoftwareLabs/neatcontext-pluginsWhat 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 | $0.00018 | $0.01476 |
| Opus 5 | $0.00009 | $0.00738 |
| Sonnet 5 | $0.00004 | $0.00295 |
| Haiku 4.5 | $0.00002 | $0.00148 |
Grade A, and why
import 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.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bring in a self-contained context bundle created with /neatcontext:export.
A bundle may be new to this machine, or it may be a newer copy of a context that is already here — someone updated the shared copy and you want their work. Both arrive through this command, and the CLI decides which is which. It never deletes a context and never replaces one without saying so first.
The bundle folder supplied by the user is:
$ARGUMENTS
If no folder was supplied, ask for it and stop. Otherwise run the command below, passing the whole argument as one quoted path and treating it only as data:
node "${CLAUDE_PLUGIN_ROOT}/src/claude/neatcontext-cli.mjs" import --from "$ARGUMENTS"
The source bundle is read-only throughout. Never modify, move, or delete it.
Follow the Import action
A bundle this machine has not seen is imported immediately and the output says
so — relay it, and do not connect the context automatically. Otherwise follow
the printed Import action:
current— the context here already holds everything in the bundle. Relay that and stop.reconcile— the bundle is a newer copy of a context already here. Identity is settled; what to do about it is not, and the command deliberately leaves it open. Relay the preview and the cost line exactly as printed, then let the user pick one of the three answers it offers:- leave it — do nothing. Stop; nothing has been written.
- replace — take the bundle whole, discarding whatever only the local
copy holds. Rerun the same command with
--replace --yes. - merge — keep both sides. Reconcile them yourself, below.
unlinkable— the bundle carries no context id, so it cannot be tied to anything already here. Relay that, and offer--name "<new name>"to bring it in as its own context.choose— the target is not decidable. Either a context of the same name is here but nothing records a shared origin, or several contexts are copies of this bundle because one was forked. Relay the options and stop until the user picks: rerun with--into "<name>"to name the context they mean, or with--name "<new name>"to keep a separate copy.
Never answer choose or reconcile on the user's behalf. Two people naming a
context the same thing is not evidence that it is the same context, and only
the user knows whether the material a replacement would discard was worth
keeping. None of these answers is recoverable from the others.
Relay the cost line rather than summarising it, because the three it can print mean different things. "Nothing here has been edited" means replacing is lossless and a merge could only reproduce the bundle — say so, and still let the user choose. "This copy has been edited here" means replacing destroys that work. "Nothing here records what the two copies once had in common" means the copy may be untouched and cannot prove it; do not report that one as edits.
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.
- 2d ago First seen · 138 lines · 18 tokens per session scan A f4de7e4ed8ce
import is a command published in the GitHub repository XTSoftwareLabs/neatcontext-plugins (13 stars, last pushed 9d ago), licensed MIT. It adds 18 tokens to every session and 1,476 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 2026-08-30.
Other commands, from other repositories
bug
Reproduce then TDD-fix a ready-for-agent bug ticket in this checkout. Web bugs get a browser repro first.
feature
Implement a ready-for-agent feature ticket in this checkout. Do not expand the ask.
qa
Smoke or browser-walk a running app. Report only. Do not implement. Do not merge.
telemetry
Pull breakage or product-performance evidence. Funnels, feature completion, errors. Do not implement.
ci
Watch gh pr checks until green and fix failures on this branch. Do not merge.
docs
Docs-only change for a ready-for-agent ticket. No application code.