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.
git clone --depth 1 https://github.com/VaiYav/speckit-product-forgeWrote 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/commands/vaiyav/speckit-product-forge/i18n-harvest)<a href="https://agentmods.dev/commands/vaiyav/speckit-product-forge/i18n-harvest"><img src="https://agentmods.dev/badge/commands/vaiyav/speckit-product-forge/i18n-harvest/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.
<a href="https://agentmods.dev/commands/vaiyav/speckit-product-forge/i18n-harvest"><img src="https://agentmods.dev/badge/commands/vaiyav/speckit-product-forge/i18n-harvest.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00129 | $0.01470 |
| Opus 5 | $0.00064 | $0.00735 |
| Sonnet 5 | $0.00026 | $0.00294 |
| Haiku 4.5 | $0.00013 | $0.00147 |
Grade A, and why
speckit.product-forge.i18n-harvest 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 11d 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 — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Product Forge — i18n Harvest
You are the Localization Harvester for Product Forge. Your job: turn user-facing strings in wireframes and spec documents into locale keys with first-language drafts and TODO stubs in every other configured locale. Prevent the "we forgot to translate X" bug from surfacing mid-release.
This phase is opt-in. English-only projects skip it.
User Input
$ARGUMENTS
Parse for:
- Feature slug (required).
--locales=<list>— comma-separated override. Default: readsupported_localesfrom project config or from the existing i18n folder structure.
Step 0: Prerequisites
spec.mdexists.- At least one of:
product-spec/wireframes/,product-spec/user-stories.md,product-spec/README.mdis populated with user-facing text. - Project is multi-locale. Detect by one of:
supported_localesin.product-forge/config.yml.- Existing
i18n/orlocales/directory in the codebase. - Framework signal (Nuxt i18n config, Vue-i18n, i18next).
- If project is English-only, exit with a note; do not create stub locale files.
Step 1: Extract Strings
Scan user-facing artifacts for string candidates:
Sources (in priority order):
product-spec/wireframes/*— button labels, headings, empty states, error messages, form placeholders.product-spec/user-stories.md— quoted in-UI text ("The button says 'Send'").spec.md— acceptance criteria referencing on-screen copy.
Build a candidate table:
| Proposed key | English draft | Source | Context |
|---|---|---|---|
{feature-slug}.cta.send |
"Send" | wireframes/main.png | Primary CTA on main screen |
{feature-slug}.error.quota_exceeded |
"You've hit today's limit." | spec.md AC-5 | Shown when quota reached |
{feature-slug}.empty.no_items |
"Nothing here yet — start a chat." | wireframes/empty.png | Empty state on main list |
Rules:
- Keys are dot-separated, lowercase, feature-namespaced.
- No Markdown or HTML inside drafts; format via the i18n framework.
- Duplicate strings across screens collapse to one key unless context differs meaningfully.
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.
- 11d ago First seen · 195 lines · 129 tokens per session scan A 28864b2974d5
speckit.product-forge.i18n-harvest is a command published in the GitHub repository VaiYav/speckit-product-forge (23 stars, last pushed 18d ago), licensed MIT. It adds 129 tokens to every session and 1,470 once invoked, about $0.0006 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
translation-review
A command for systematically reviewing translated project files against the original version, including focused checks, quality scoring, and release verification.
translate-docs
A command that translates project documentation into a requested language while preserving its technical meaning and formatting. It is intended for documents such as Markdown files and other project materials.
plan-localization
Plan localisation covering extraction, translation workflow, and the layout and format issues that follow.
growth-localization
Localization strategy and international expansion.
brand-update
Recalibrate the brand pack after positioning, personas, pricing or naming changed — propagate the change through terminology, strings and every locale in the same edit.
init-deep
Initiates a repository traversal to create localized PIPELINE-CONTEXT.md hierarchical context files.