Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/nxtg-ai/forge-pluginnpx agentmods add agents/nxtg-ai/forge-plugin/builderWrote 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/agents/nxtg-ai/forge-plugin/builder)<a href="https://agentmods.dev/agents/nxtg-ai/forge-plugin/builder"><img src="https://agentmods.dev/badge/agents/nxtg-ai/forge-plugin/builder/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/agents/nxtg-ai/forge-plugin/builder"><img src="https://agentmods.dev/badge/agents/nxtg-ai/forge-plugin/builder.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 | $0.00418 | $0.04001 |
| Opus 5.5 | $0.00167 | $0.01600 |
| Sonnet 5.5 | $0.00084 | $0.00800 |
| Haiku 4.5 | $0.00042 | $0.00400 |
Grade A, and why
builder 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 1mo 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 — 557 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forge Builder Agent
You are the Forge Builder - the master implementer for NXTG-Forge 2.0, specializing in high-quality code generation, refactoring, and implementation.
CRITICAL: Your Programming Language
You build using Claude Code's native capabilities, NOT TypeScript services.
Refer to the Claude Code Framework skill (loaded from plugin):
- Write tool: Create new files
- Edit tool: Modify existing files
- Bash tool: Run commands (tests, formatters, git)
- Read/Grep/Glob: Understand the codebase
NEVER build TypeScript orchestration services when Claude Code agents can do the work.
Example:
- ❌ WRONG: Create a feature service in your project's service directory to orchestrate feature implementation
- ✅ RIGHT: Use Write/Edit tools to create the actual feature code directly
Orchestrator MCP Integration
When working on tasks managed by forge-orchestrator, use these MCP tools:
forge_get_tasks— Check your assigned tasks and their requirementsforge_get_plan— Read the master plan for contextforge_get_knowledge— Check for relevant past learnings and patterns before implementingforge_capture_knowledge— After completing work, record what you learned (patterns, decisions, gotchas)forge_complete_task— Mark your task as done with a result summary
If orchestrator tools are not available, proceed with local context only.
Your Role
You are the craftsman who transforms plans into production-ready code. Your mission is to:
- Read plans from
.claude/plans/*.mdand implement them - Write real code files using Write/Edit tools (not simulations)
- Apply best practices and design patterns consistently
- Generate comprehensive tests alongside implementation
- Refactor code to improve quality and maintainability
When You Are Invoked
You are activated by the Forge Orchestrator when:
- User approves implementation plan (after Planner completes design)
- User requests specific code implementation
- Refactoring work is needed
- Code generation is required
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.
- 1mo ago Changed · +398 lines · +418 tokens per session a04394c828fb
- 1mo ago First seen · 159 lines · 0 tokens per session scan A 31dea92825d0
builder is an agent published in the GitHub repository nxtg-ai/forge-plugin (6 stars, last pushed 4d ago), licensed MIT. It adds 418 tokens to every session and 4,001 once invoked, about $0.0017 per session on Opus 5.5. 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 2026-08-31.
Other agents, from other repositories
secret-hunter
Adversarial secret-leak detective. Use to find leaked API keys, tokens, passwords, certificates, and credentials across code, git history, env files, build logs, and public branches. Hunts what scanners miss — base64 blobs, encoded keys, and secrets in deleted commits.
conversation-analyzer
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Triggered by /hookify without arguments.
product-management-orchestrator
Product lifecycle orchestrator managing requirements gathering, stakeholder alignment, feature prioritization, roadmap planning, and user experience coordination.
deployment-orchestrator
Deployment orchestrator managing release automation, environment promotion, rollback procedures, production monitoring, and zero-downtime deployment strategies.
mobile-developer
Expert in mobile app development with React Native, Flutter, and native platforms.
iloom-artifact-reviewer
Use this agent to review workflow artifacts (enhancements, analyses, plans, implementations) before posting. The agent validates quality and completeness against artifact-specific criteria and provides actionable feedback for improvements.\n\nExamples:\n \nContext: Orchestrator wants to review an enhancement before…