By tqnonline
Intelligent prompt engineering suite with signal-based technique selection (10 techniques across 3 tiers), automatic hardening patterns (10 composable patterns), UNIX-philosophy orchestrator agent, and voice calibration. Transforms raw intent into production-grade AI prompts.
Intelligent prompt engineering orchestrator agent. The single entry point for building high-quality, context-aware AI prompts. Unlike the prompt-forge skill (which follows a fixed 5-phase dialogue), this agent comprehends context autonomously, asks minimal smart questions, auto-selects techniques and hardening patterns, and assembles production-grade prompts. Follows UNIX philosophy: small composable stages, text in / text out, do one thing well at each stage. TRIGGER when: user says "prompt-forge-agent", "build me a smart prompt", "forge agent", "help me prompt", "craft an agent prompt", or describes a complex task that needs an optimized prompt with automatic technique selection. Routes to: prompt-forge (skill execution), humanize (voice calibration for artifacts).
Intelligent prompt engineering orchestrator agent. The single entry point for building high-quality, context-aware AI prompts. Unlike the prompt-forge skill (which follows a fixed 5-phase dialogue), this agent comprehends context autonomously, asks minimal smart questions, auto-selects techniques and hardening patterns, and assembles production-grade prompts. Follows UNIX philosophy: small composable stages, text in / text out, do one thing well at each stage. TRIGGER when: user says "prompt-forge-agent", "build me a smart prompt", "forge agent", "help me prompt", "craft an agent prompt", or describes a complex task that needs an optimized prompt with automatic technique selection. Routes to: prompt-forge (skill execution), humanize (voice calibration for artifacts).
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A curated, BSD-3-Clause marketplace of agent skills, plugins, and slash commands. Works natively with Claude Code, GitHub Copilot CLI, OpenAI Codex CLI, Cursor, and Factory.ai Droid (CLI + desktop app). Git-URL install for Amp and Gemini CLI. Copy-to-home adapters for Kilo Code, OpenCode, and Crush. Portable prompt loaders for Perplexity, ChatGPT GPTs, and Claude.ai Projects.
13 install targets. 4 plugins. 46 skills. One canonical source.
Claude Code:
claude plugin marketplace add github:tqnonline/agent-forge
claude plugin install writing
GitHub Copilot CLI:
copilot plugin marketplace add tqnonline/agent-forge
Codex CLI:
codex plugin marketplace add tqnonline/agent-forge
Cursor: Open cursor.com/marketplace, search "agent-forge", click Install.
Factory.ai Droid (CLI + desktop):
droid plugin marketplace add https://github.com/tqnonline/agent-forge
droid plugin install writing@agent-forge
Amp:
amp skill add github:tqnonline/agent-forge
Gemini CLI:
gemini skills install github:tqnonline/agent-forge
Kilo Code, OpenCode, or Crush:
pipx install tqn-agent-forge # or: uv tool install tqn-agent-forge
agent-forge install writing --tier kilocode # or opencode, or crush
Perplexity, ChatGPT GPTs, Claude.ai Projects: See docs/install/_index.md — these use prompt loaders rather than native install.
| Plugin | Skills | Description |
|---|---|---|
writing | humanize | Detect AI writing patterns and rewrite in natural human voice |
prompts | prompt-forge | Interactive prompt engineering via structured dialogue |
msft-arch | 32 skills | Microsoft enterprise architecture: Azure, M365, Power Platform, Dynamics 365 |
pm | 12 skills | Product/Program Management: PRD generation, epic decomposition, 11-Star review |
Paste this into any LLM agent:
Install agent-forge plugins from
https://raw.githubusercontent.com/tqnonline/agent-forge/main/docs/install/_index.md
The agent fetches the catalog, identifies your tool, and runs the correct install commands.
agent-forge update --check # see what's stale
agent-forge update # apply
See CONTRIBUTING.md. Three paths: typo fix (no issue needed),
new skill (must include eval suite), new plugin (requires ADR discussion). DCO sign-off
required (git commit -s).
BSD-3-Clause. Per-plugin assets may carry their own licenses; see THIRD_PARTY_NOTICES.md.
npx claudepluginhub tqnonline/agent-forge --plugin promptsCollection of writing transformation and text processing skills including humanization, voice calibration, and AI pattern detection
Product and Program Management skill suite for technology industries. 12 composable skills following UNIX philosophy: philosophy, discover, map, tom-architect, epic-decompose, prd-draft, prd-validate, prd-review, backlog-decompose, backlog-ado, backlog-linear, backlog-github. Includes Product Constitution (living strategy doc), multi-platform TOM design, and 11-Star PRD quality framework. Works for SaaS product companies and consulting delivery firms.
Microsoft enterprise architecture suite: 32 specialist skills across 5 tiers (process, verticals, horizontals, long-tail specialists, output, cross-cutting). FINAL v5 expansion complete. Vertical specialists (Azure, Power Platform, D365, Containers, Data, AI) build solutions; horizontal architects (Identity, Security, IaC, .NET, FinOps, Observability, CI/CD, Threat Model) review them in parallel; long-tail specialists (MAUI, SRE, .NET Modernization, Defender+Sentinel) cover Phase 4 domains. Always-on horizontals run after every vertical; trigger-based and stack-pinned routing fire on signal match. Odin resolves cross-horizontal conflicts. Validate runs Council of Three on every output package.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Write feature specs, plan roadmaps, and synthesize user research faster. Keep stakeholders updated and stay ahead of the competitive landscape.
Curate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.