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/ganbin/4d-development-skill/dialoggit clone --depth 1 https://github.com/Ganbin/4d-development-skillWrote 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/ganbin/4d-development-skill/dialog)<a href="https://agentmods.dev/commands/ganbin/4d-development-skill/dialog"><img src="https://agentmods.dev/badge/commands/ganbin/4d-development-skill/dialog.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 | $0.02210 |
| Opus 5 | $0.00000 | $0.01105 |
| Sonnet 5 | $0.00000 | $0.00442 |
| Haiku 4.5 | $0.00000 | $0.00221 |
Grade A, and why
dialog 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 today.
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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
| Parameter | Type | Description | |
|---|---|---|---|
| aTable | Table | → | Table owning the form or If omitted: default table or use of project form |
| form | Text, Object | → | Name (string) of table or project form, or a POSIX path (string) to a .json file describing the form, or an object describing the form |
| formData | Object | → | Data to associate to the form |
| * | Operator | → | Use the same process |
Description
This command is designed to work with customized and advanced user interfaces based on forms. You can use it to display information coming from the database or other locations, or to provide data entry features. Unlike ADD RECORD or MODIFY RECORD, DIALOG gives you full control over the form, its contents and the navigation and validation buttons.
This command is typically called along with the Open form window to display sophisticated forms, as shown in the following example:

The DIALOG command can also be used instead of ALERT, CONFIRM or Request when the information to be presented or gathered is more complex than those commands can manage.
In the form parameter, you can pass:
- the name of a form (project form or table form) to use;
- the path (in POSIX syntax) to a valid .json file containing a description of the form to use;
- an object containing a description of the form to use.
Optionally, you can pass parameter(s) to the form using a "form data" object. Any properties of the form data object will then be available from within the form context through the Form command. For example, if you use a form data object containing {"version";"12"}, you will be able to get or set the value of the "version" property in the form by calling:
$v:=Form.version //"12"
Form.version:=13
To fill the "form data" object, you have two possibilities:
-
use the formData parameter. Using a local variable for formData allows you to safely pass parameters to your forms, whatever the calling context. In particular, if the same form is called from different places in the same process, you will always be able to access its specific values by simply calling Form.myProperty. Moreover, since objects are passed by reference, if the user modifies a property value in the form, it will automatically be saved in the object itself.
-
associate a user class to the form, in which case 4D will automatically instantiate an object of this class when the form will be loaded. The object properties and functions will be automatically available through the object returned by Form. You could write for example
Form.myFunction().
:::note
- The formData parameter has priority over a form class (the class object is not instantiated if a formData parameter is passed).
- If you do not pass the formData parameter (or if you pass an undefined object) and no user class is associated to the form, DIALOG creates a new empty object bound to the form.
:::
The dialog is closed by the user either with an "accept" action (triggered by the ak accept standard action, the Enter key, or the ACCEPT command), or with a "cancel" action (triggered by the ak cancel standard action, the Escape key, or the CANCEL command). An accept action will set the OK system variable to 1, while a cancel action will set OK to 0.
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.
- today First seen · 181 lines · 0 tokens per session scan A 367d864d262f
dialog is a command published in the GitHub repository Ganbin/4d-development-skill (9 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,210 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 2026-09-03.
Other commands, from other repositories
doctor
Run a system health check to diagnose common issues.
completions
Generate shell completion scripts for bash, zsh, and fish.
watch
Watch installed skills for file changes and auto-sync to agent directories.
upgrade
Upgrade rolecraft itself to the latest version.
querying
Query documents from a search index using type-safe filters with support for pagination, sorting, field selection, scoring, and highlighting. Count matching documents efficiently without returning results.
automode-edit
Edit the project autoMode block in plain language. The agent interprets the request, builds a proposal, runs critique + hash gate, and writes atomically to .claude/settings.local.json.