{"id":29998137,"url":"https://github.com/othercodes/minion-forge","last_synced_at":"2026-02-11T19:05:17.819Z","repository":{"id":305997111,"uuid":"1024570844","full_name":"othercodes/minion-forge","owner":"othercodes","description":"MinionForge is a modular instruction system for AI models. It is designed to work with Claude Sonnet 4, GPT-based models, or any language model that supports XML-style prompts inside Markdown files.","archived":false,"fork":false,"pushed_at":"2025-07-24T08:08:12.000Z","size":33,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-08-05T03:41:43.269Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/othercodes.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-07-22T23:19:06.000Z","updated_at":"2025-07-25T08:46:55.000Z","dependencies_parsed_at":"2025-07-23T05:34:32.399Z","dependency_job_id":null,"html_url":"https://github.com/othercodes/minion-forge","commit_stats":null,"previous_names":["othercodes/minion-forge"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/othercodes/minion-forge","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/othercodes%2Fminion-forge","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/othercodes%2Fminion-forge/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/othercodes%2Fminion-forge/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/othercodes%2Fminion-forge/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/othercodes","download_url":"https://codeload.github.com/othercodes/minion-forge/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/othercodes%2Fminion-forge/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29341724,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-11T18:58:20.535Z","status":"ssl_error","status_checked_at":"2026-02-11T18:56:44.814Z","response_time":97,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-08-05T03:41:15.996Z","updated_at":"2026-02-11T19:05:17.804Z","avatar_url":"https://github.com/othercodes.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# MinionForge\n\nMinionForge is a modular instruction system for AI models. It is designed to work with Claude Sonnet 4, GPT-based models, or any language model that supports XML-style prompts inside Markdown files.\n\nThe goal is to define precise, reusable, and role-based behaviors through structured prompt modules. This helps AI follow clear rules and adapt better to real-world project workflows.\n\n## Motivation\n\nMost prompts are plain text — difficult to maintain, scale, or reuse reliably.\n\nMinionForge solves this by breaking behavior into modular, reusable instructions tied to specific roles and contexts.\n\n## Project Status\n\nMinionForge is currently in active development and considered experimental. The format, tags, and conventions may change as we refine the design based on real-world usage with different models.\n\n## References\n\n- [Anthropic Claude Prompt Engineering Guide](https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/overview)\n\n## Components\n\nMinionForge uses custom XML tags embedded in `.md` files. These tags define structure and intent, while keeping the text readable and easy to maintain.\n\n| Tag             | Purpose                                                           |\n|------------------|--------------------------------------------------------------------|\n| `\u003crole\u003e`         | Declares the identity and purpose of the AI agent.                |\n| `\u003cresponsibility\u003e` | Lists what this role is expected to deliver.                    |\n| `\u003cdefault-behavior-rules\u003e` | Declares global rules shared by all roles and instruction. Only allowed in the base instruction file. |\n| `\u003crules\u003e` / `\u003crule\u003e` | Mandatory behaviors and constraints for the role or instruction. |\n| `\u003cdirectives\u003e` / `\u003cdirective\u003e` | Recommended strategies or behavioral suggestions.      |\n| `\u003cinstruction\u003e` / `\u003cinstructions\u003e` | Declares instructions: either as inline blocks or external references. |\n| `\u003cexamples\u003e` / `\u003cexample\u003e` | Optional usage samples for clarity or guidance.          |\n| `\u003cswitches\u003e` / `\u003cswitch\u003e` | Runtime options or behavior flags (e.g. `--seq`).         |\n| `\u003ccontext-binding\u003e` | Sets the default language, architecture, or project context.   |\n| `\u003ccommunication-rules\u003e` | Defines how the AI should interact with users (tone, clarity, intent). |\n| `\u003ctone-rules\u003e`          | Enforces voice, confidence, and level of formality.                  |\n| `\u003cvisual-style-rules\u003e`  | Controls formatting, spacing, line length, and visual hierarchy.     |\n| `\u003csymbol-rules\u003e`        | Specifies allowed symbols and their meanings.                        |\n| `\u003csymbol\u003e`              | Declares individual symbols and their interpretation.                |\n| `\u003cmessage-structure\u003e`   | Defines the required structure of output (summary first, bullets, etc). |\n\n## Format and Structure\n\nAll files are plain `.md` files using XML-style tags. Each `\u003cinstruction\u003e` is structured using a consistent set of components described below.\n\n### `\u003cinstructions\u003e` / `\u003cinstruction\u003e`\n\nThe `\u003cinstructions\u003e` tag is used to declare atomic instructions.\n\nYou can define them in two ways:\n\n1. **External reference** (recommended for modular design):\n\n```xml\n\u003cinstructions\u003e\n  - [analyze-project](../instructions/analyze-project.md)\n  - [generate-development-checklist](../instructions/generate-development-checklist.md)\n\u003c/instructions\u003e\n```\n\neach item links to a separate `.md` file containing a \u003cinstruction\u003e block, like:\n\n```xml\n\u003c!-- File: instructions/generate-development-checklist.md --\u003e\n\u003cinstruction id=\"generate-development-checklist\"\u003e\n  Generate a detailed checklist from the FDD, with verifiable implementation steps grouped by section.\n\u003c/instruction\u003e\n```\n\nYou may also define a natural-language condition inline to guide activation:\n\n```xml\n\u003cinstructions\u003e\n  Each listed instruction is REQUIRED when the user’s request matches the described condition.\n  You MUST load and apply the full content of the linked file in that case.\n\n  - You MUST use [generate-tests](./instructions/generate-tests.md) whenever the user requests to create, extend, or infer unit or feature tests from code or specification.\n\u003c/instructions\u003e\n```\n\nThis is an example of a **semantic activation phrase**. It tells the model when to apply each instruction, using natural language instead of special tags. This approach also makes the instruction list easier to read and maintain.\n\n2. **Inline declaration** (for context-specific or temporary instructions):\n\n```xml\n\u003cinstructions\u003e\n  \u003cinstruction id=\"generate-development-checklist\"\u003e\n    Generate a detailed checklist from the FDD, with verifiable implementation steps grouped by section.\n  \u003c/instruction\u003e\n\u003c/instructions\u003e\n```\n\nYou may mix both forms if needed, but external files are preferred for reuse and clarity.\n\n### `\u003crules\u003e`\n\nDefines strict behaviors or constraints the model must follow.  \nEach `\u003crule\u003e` is atomic and imperative. Use them to enforce logic, compliance, or limitations.\n\nYou can add a `for=\"...\"` attribute to scope the rule (e.g., `tooling`, `architecture`, etc.).\n\n```xml\n\u003crules\u003e\n  \u003crule\u003eNEVER write business logic inside views or serializers.\u003c/rule\u003e\n  \u003crule\u003eALWAYS validate user input before processing.\u003c/rule\u003e\n  \u003crule for=\"tooling\"\u003eONLY use markdown for documentation tasks.\u003c/rule\u003e\n\u003c/rules\u003e\n```\n\n### `\u003cdirectives\u003e`\n\nProvides softer behavioral instructions — best practices, style guides, or process recommendations.  \nUse `\u003cdirective\u003e` for suggestions that improve clarity, safety, or quality.\n\n```xml\n\u003cdirectives\u003e\n  \u003cdirective\u003eUse second-level markdown headings (`##`) for main sections.\u003c/directive\u003e\n  \u003cdirective\u003eOrganize tasks in the same order as the FDD sections.\u003c/directive\u003e\n  \u003cdirective\u003eAsk questions before adding optional sections like security or performance.\u003c/directive\u003e\n\u003c/directives\u003e\n```\n\n### `\u003cexamples\u003e`\n\nOptional but useful. Provide sample inputs and expected outputs using `\u003cexample\u003e` to guide the model.  \nUse only when the behavior is non-obvious or benefits from demonstration.\n\n```xml\n\u003cexamples\u003e\n  \u003cexample\u003e\n    \u003cinput\u003eDesign an endpoint to deactivate a user.\u003c/input\u003e\n    \u003coutput\u003ePOST /users/{id}/deactivate → triggers UserDeactivationService\u003c/output\u003e\n  \u003c/example\u003e\n\u003c/examples\u003e\n```\n\n### `\u003cswitches\u003e`\n\nDeclare runtime flags that affect reasoning. For example: `--seq` for sequential thinking.  \nEach `\u003cswitch\u003e` represents a modifier that the model should activate when the instruction is loaded.\n\n```xml\n\u003cswitches\u003e\n  \u003cswitch\u003e--seq\u003c/switch\u003e\n  \u003cswitch\u003e--context7\u003c/switch\u003e\n\u003c/switches\u003e\n```\n\n### `\u003cvariants\u003e`\n\nUsed to specialize a instruction for specific environments (like Django, Laravel, etc.).\n\nEach `\u003cvariant\u003e` must include a `\u003ccontext-match\u003e` section that defines when it applies.  \nVariants may override or add `\u003crules\u003e`, `\u003cdirectives\u003e`, or `\u003cexamples\u003e` inside the instruction.\n\n```xml\n\u003cvariant\u003e\n  \u003ccontext-match\u003e\n    \u003clanguage\u003ePython\u003c/language\u003e\n    \u003cframework\u003eDjango\u003c/framework\u003e\n  \u003c/context-match\u003e\n  \u003crules\u003e\n    \u003crule\u003eUse DRF ViewSets and Celery tasks when applicable.\u003c/rule\u003e\n  \u003c/rules\u003e\n\u003c/variant\u003e\n```\n\nVariants are optional. The base instruction applies globally by default.\n\n### `\u003ccontext-binding\u003e`\n\nDeclares the default working context — such as language, framework, architecture, constraints, or any other environment-specific assumption the AI should consider when reasoning.\n\nUse this block in role or system instruction files to define expectations like:\n\n```xml\n\u003ccontext-binding\u003e\n  \u003clanguage\u003ePython\u003c/language\u003e\n  \u003cframework\u003eDjango\u003c/framework\u003e\n  \u003cquality\u003emodular, testable\u003c/quality\u003e\n  \u003cprinciples\u003edesign-first\u003c/principles\u003e\n\u003c/context-binding\u003e\n```\n\n### Example Project Structure\n\n```text\n.github/\n  ├── copilot-instructions.md\n  ├── instructions/\n  │   ├── analyze-project.md\n  │   ├── generate-fdd-outline.md\n  │   ├── generate-development-checklist.md\n  │   ├── generate-architecture-diagram.md\n  │   └── ...\n  └── chatmodes/\n      ├── feature-designer.chatmode.md\n      └── software-engineer.chatmode.md\n\n```\n\n- `instructions/` contains atomic instruction modules (reusable).\n- `chatmodes/` defines the roles, responsibilities, and instruction loading logic.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fothercodes%2Fminion-forge","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fothercodes%2Fminion-forge","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fothercodes%2Fminion-forge/lists"}