A guide for extending Amy, the Amethyst command-line tool, including adding commands and deciding which project module should contain new logic. A command-line tool is a program operated through a terminal.

Use it when adding an Amy subcommand, editing its CLI code, writing command-line integration tests, or moving shared logic into the correct Amethyst module.

It helps prevent the CLI from becoming a place for business logic or protocol code, which would make the project harder to maintain and could break its public command interface.

How vitorpamplona/amethyst configures Claude Code

About the project

Amethyst is an Android client for Nostr, a social network protocol that lets people control their own social activity and connections. People use it to read and publish Nostr content, follow accounts, and exchange encrypted direct messages from Android and other supported platforms. The catalogue add-ons support workflows for developing or operating the client.

vitorpamplona/amethyst · 1,620 stars · on GitHub · amethyst.social

Reuse

Borrowing it

Nothing to install: this file belongs to vitorpamplona/amethyst. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/vitorpamplona/amethyst/main/.claude/skills/amy-expert/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/vitorpamplona/amethyst

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/vitorpamplona/amethyst/amy-expert/github.svg)](https://agentmods.dev/skills/vitorpamplona/amethyst/amy-expert)

Your own site

<a href="https://agentmods.dev/skills/vitorpamplona/amethyst/amy-expert"><img src="https://agentmods.dev/badge/skills/vitorpamplona/amethyst/amy-expert/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/vitorpamplona/amethyst/amy-expert"><img src="https://agentmods.dev/badge/skills/vitorpamplona/amethyst/amy-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,543 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

high Privilege Escalation · line 175
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
medium Rogue Agent · line 94
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

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	$0.00000	$0.01543
Opus 5.5	$0.00000	$0.00617
Sonnet 5.5	$0.00000	$0.00309
Haiku 4.5	$0.00000	$0.00154

Measured 25d ago against content hash 15ed426ea003, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

amy-expert 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 25d 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.

.claude/skills/amy-expert/SKILL.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

amy CLI

cli/ is a non-interactive JVM client over quartz + commons. Its --json output and exit codes are a public API that the interop harnesses depend on. The long-form contract is cli/DEVELOPMENT.md ("Public contract", "How to add a command", "Output-shape conventions", "Testing", "Full on-disk layout"); this skill is the short version plus the traps.

Rules

Thin assembly layer. No protocol, filter assembly, state machines or encryption in cli/. Protocol goes to quartz/; business logic to commons/, extracted from amethyst/ first in its own commit if needed (references/extraction-recipe.md). A commands/*.kt file over ~200 lines is a smell. cli may depend on :geode (for amy serve), never on :commonsUI, :amethyst or :desktopApp.
Output goes through Output (cli/src/main/kotlin/…/cli/Output.kt).
- Output.emit(mapOf(...)): the map is the --json shape (one line, one object, stable snake_case keys); the default text render comes from the same map and has no contract.
- return Output.error("code", "detail"): the return value is the exit code (bad_args 2, timeout 124, anything else 1). Errors go to stderr (error: code: detail, or a JSON object under --json).
- Exit codes 0 ok, 1 runtime, 2 bad args, 124 await timeout. runCli in Main.kt also maps a thrown IllegalArgumentException to 2 and AwaitTimeout to 124.
- Never println results; System.err.println is for progress logs only.
- Adding a --json key is safe. Renaming or removing one is breaking: say so in the commit, and update cli/README.md and the fixtures under cli/tests/.
- Publishes: RawEventSupport.publishGuard(ack, id) then + RawEventSupport.ackFields(ack) (published_to + rejected_by: [{relay, reason}]). Don't hand-roll the projection.
- Reuse an error code from the canonical list in cli/DEVELOPMENT.md before minting one.
Non-interactive. Everything is a flag; a network wait is an explicit await verb with --timeout. The only TTY fallback is PassphraseProvider (after --passphrase-file and $AMY_PASSPHRASE); don't add more.
~/.amy/ is the whole state, reloaded on every run; no in-process caches across runs. ~/.amy/shared/events-store/ (one event store for all accounts), ~/.amy/<account>/ (identity.json, state.json, aliases.json, marmot/), ~/.amy/current (written by amy use NAME). Account selection is the global --account NAME; --name stays free for subcommands. New persisted state goes through Config.kt / stores/FileStores.kt / a named helper like Aliases.kt.
$HOME, not user.home. DataDir.DEFAULT_ROOT (Config.kt) reads $HOME; JDK 21's user.home comes from getpwuid and ignores it, which would break harness isolation (HOME=$(mktemp -d) amy --account alice init). JVM tests use the amy.home system property seam instead.

Read the full file on GitHub · 80 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

references/extraction-recipe.md 5.6 KB

Changes

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.

25d ago Changed · -198 lines · -217 tokens per session 15ed426ea003
1mo ago Changed · +19 lines 24f8d6eb28ed
2mo ago First seen · 259 lines · 217 tokens per session scan A 3cb364436e40

Subscribe to this mod's changes

amy-expert is a skill published in the GitHub repository vitorpamplona/amethyst (1,620 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,543 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.

Kotlin and Swift Languages android kotlin nostr social-network

amy-expert next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
amy-expertvitorpamplona/amethyst	Skill	1.6k	yesterday	MIT	A	0	Claude Code
compose-animationschrisbanes/skills	Skill	1.1k	yesterday	Apache-2.0	A	64	Claude Code
compose-focus-navigationchrisbanes/skills	Skill	1.1k	yesterday	Apache-2.0	A	41	Claude Code
kotlin-androidericrisco/rsc-harness	Skill	503	2d ago	MIT	A	78	Claude Code, Codex
compose-multiplatformericrisco/rsc-harness	Skill	503	2d ago	MIT	A	80	Claude Code, Codex
release-kotlin-librarychrisbanes/skills	Skill	1.1k	yesterday	Apache-2.0	A	46	Claude Code
using-chrisbanes-skillschrisbanes/skills	Skill	1.1k	yesterday	Apache-2.0	A	38	Claude Code

amethyst: Skill for Claude Code