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/korjwl1/toki/releasegit clone --depth 1 https://github.com/korjwl1/tokiWhat 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 | $0.00000 | $0.01135 |
| Opus 5 | $0.00000 | $0.00567 |
| Sonnet 5 | $0.00000 | $0.00227 |
| Haiku 4.5 | $0.00000 | $0.00113 |
Grade A, and why
release 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Release a new version of toki. macOS binaries are built locally, Linux binaries are built via GitHub Actions.
Instructions
When the user runs /release, follow these steps:
1. Determine version
- Ask the user for the version to release (e.g.,
2.3.0) - Or if the user already provided it as an argument (e.g.,
/release 2.3.0), use that - Validate it follows semver format (MAJOR.MINOR.PATCH or MAJOR.MINOR.PATCH-prerelease)
2. Pre-flight checks
- Run
cargo testto make sure tests pass - Run
cargo clippy --all-targets --all-features -- -D warnings. The current development branch has a warning baseline, so this strict release gate is expected to block until those warnings are resolved; do not mistake plain Clippy's exit 0 for a clean release check. - Check
git statusto ensure the working tree is clean (no uncommitted changes) - Check that the tag
v{VERSION}doesn't already exist:git tag -l v{VERSION} - Confirm
Cargo.tomlhas no local[patch."https://github.com/korjwl1/toki-sync-protocol.git"]section. A tagged release/Actions checkout has no sibling../toki_sync_protocol, so releasing with the patch is guaranteed to fail. - Confirm the protocol dependency is pinned to a published tag containing every wire type the client uses. The current baseline is
toki-sync-protocolv1.1.0; releasetoki-syncbefore consumers when a future protocol change adds server capabilities.
If any check fails, report the issue and stop.
3. Update version
- Update
versioninCargo.tomlto the new version - Run
cargo checkto updateCargo.lock - Commit with message:
chore: bump version to {VERSION}
4. Build macOS targets (local)
cargo build --release --target aarch64-apple-darwin
cargo build --release --target x86_64-apple-darwin
Package into tar.gz:
for target in aarch64-apple-darwin x86_64-apple-darwin; do
tar -czf "toki-{VERSION}-${target}.tar.gz" -C "target/${target}/release" toki
done
5. Create tag, push, and upload
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 · 100 lines · 0 tokens per session scan A 17a4a588d662
release is a command published in the GitHub repository korjwl1/toki (21 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,135 tokens. 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
tokenless-stats
Show Tokenless compression statistics.
stats
Show Claude Code spending analytics (today/week/month/all/tag/branch).
run
显式启动一次隔离的 Windows Computer Use 任务.
test
运行 Computer Use 通用能力测试.
by-project
Claude Code spend per project (branches merged) with today / this week / this month and the last active date, ranked by month spend. Computed locally from the session logs Claude Code already writes. Never touches API traffic; zero key, zero prompts, zero latency.
port-loop
Iterate research → implement → verify autonomously until a capability works or the model of it is exhausted (linux-phone-porting).