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 skills/anomalyco/hex/personal-commandsnpx skills add anomalyco/hex --skill personal-commandsgit clone --depth 1 https://github.com/anomalyco/hexWrote 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/anomalyco/hex/personal-commands)<a href="https://agentmods.dev/skills/anomalyco/hex/personal-commands"><img src="https://agentmods.dev/badge/skills/anomalyco/hex/personal-commands.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 | $0.00000 | $0.01779 |
| Opus 5 | $0.00000 | $0.00890 |
| Sonnet 5 | $0.00000 | $0.00356 |
| Haiku 4.5 | $0.00000 | $0.00178 |
Grade A, and why
personal-commands 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.
How it starts
The opening of the file, as written. The whole thing — 236 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Custom Commands
Edit hex.config.ts, not .hex-sdk. Config and installed dependencies execute
with the user's normal permissions; use only trusted packages and avoid network,
filesystem, environment, or subprocess access unless the requested command
requires it. Run bun run check after editing.
Spoken commands and dictation controls require the Commands opt-in, which
defaults off; a config does not enable it. Mode transformations can run with
Commands off once selected. Bun must be installed separately, and the host
requires the exact Effect version in .hex-sdk/package.json even for
Promise-only configs.
bun run check validates TypeScript, not phrase overlaps. A running host watches
regular .ts and .json workspace files, excluding node_modules, .git, and
symlink targets. Check the Commands pane for runtime reload errors. If missing
config, Bun, or dependencies prevented host startup, restart HEX after repair.
Replace the native voice-dictation phrases declaratively when desired:
import { defineHexConfig } from "@hex/commands"
export default defineHexConfig({
dictation: {
start: ["begin note"],
stop: ["finish note"],
send: ["send note"],
cancel: ["discard note"],
},
commands: {},
})
start is a streaming utterance prefix while HEX is listening. The other
controls are streaming suffixes only during an active voice capture. This block
replaces the native phrases exactly, not the capture lifecycle, and does not
define command handlers. Each list must be nonempty and unambiguous. If this
block is omitted, HEX uses its built-in protocol.
Register named finishing transformations when a dictation mode needs a deterministic text rule:
import { defineHexConfig } from "@hex/commands"
export default defineHexConfig({
transformations: {
"trim-whitespace": {
name: "Trim whitespace",
description: "Remove leading and trailing whitespace",
transform: (text) => text.trim(),
},
},
commands: {},
})
Select the registered transformation in the mode's Transformations section. Selected transformations run in displayed order after that mode's corrections and optional AI rewrite for Paste and Send, not Voice Action or meetings. Transformations may be async, receive the foreground context as their second argument, and must return a string. If the transformation stage fails, HEX keeps the text from before that stage, not partial transformation results.
Do not reuse lowercase or spongebob-case as custom transformation IDs; these
IDs execute native built-ins instead of a registered TypeScript function.
Use a handler with the provided hex capabilities for ordinary commands:
import { defineHexConfig } from "@hex/commands"
export default defineHexConfig({
commands: {
slack: {
phrases: ["open slack"],
run: ({ hex }) => hex.openApplication("Slack"),
},
},
})
To navigate inside an app, use openUrl (lowercase rl) with its deep link.
openApplication takes an app name or path, not a URL. macOS must have an
installed handler for the URL scheme:
run: ({ hex }) => hex.openUrl("slack://channel?team=T_EXAMPLE&id=C_EXAMPLE")
URLs are passed unchanged, including their query and fragment. File URLs and
inline script/data URLs are unsupported; use openPath for filesystem paths.
Handlers may issue several calls. Await dependent capability calls to preserve their order; unawaited calls can run concurrently even though HEX tracks them until handler completion:
run: async ({ hex }) => {
await hex.openApplication("Slack")
await hex.press({ key: "k", modifiers: ["command"] })
}
End a phrase with one {name} placeholder to capture the rest of the spoken
words. The normalized remainder arrives as captures.name: ASCII letters are
lowercased, ASCII punctuation is trimmed from word boundaries, whitespace is
collapsed, and spoken digits zero through nine become 0 through 9.
Capture phrases require run and at least one spoken word before the placeholder:
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 Changed · +27 lines a2194440a16d
- 5d ago First seen · 209 lines · 0 tokens per session scan A b03c1588891c
personal-commands is a skill published in the GitHub repository anomalyco/hex (118 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,779 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 skills, from other repositories
repo-build-pr
End-to-end implementation loop for the os-clovy repo: study the prompt, ask the clarifying questions that change what gets built, plan and architect on the session model while delegating bulk implementation and review to cheaper strong models, isolate work in git worktrees, validate with deterministic checks plus…
os-accounts-integration
Integrate consumer apps, Tauri desktop clients, or backend services with OS Accounts, the Open Software identity and credit-billing platform. Use when adding Login with Open Software, PKCE callback handling, /me, access-token verification with JWKS/ES256, auth middleware for Hono/Fastify/Express/Rust, top-up links…
os-rust-backend
Use when working on a Rust backend that follows the Open Software Network house style — the seven-crate Cargo workspace split (domain / services / persistence / providers / config / api / app) shared by os-accounts and os-platform (fellow). Trigger on: scaffolding a new service in this org ("like os-accounts"…
os-platform
Query and update live Open Software os-platform production data through the platform API. Use when an agent needs current Issues/Bounties, Orgs, Projects, Submissions, Comments, Activity, Contributors, Product Memory, the team Timeline, or API status, or needs to create, assign, move, or comment on tracked work.
os-rust-backend-ci
Use this skill when the user wants to set up, configure, debug, or extend a GitHub Actions pipeline for a Rust backend that runs CI checks and produces a container image artifact in GHCR. Scope is testing + artifact only — deployment is intentionally out of scope. Trigger on: PR checks (fmt, clippy, tests, OpenAPI /…
agent-e2e-qa
Run live, agent-driven end-to-end QA for os-clovy by opening the real app or web preview, clicking through changed flows, inspecting visible state, recording and compressing video, attaching QA videos to PRs through os-platform file uploads, capturing screenshots/logs, and reporting pass/fail evidence. Use when a user…