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/SteveGJones/ai-first-sdlc-practicesnpx agentmods add agents/stevegjones/ai-first-sdlc-practices/pipeline-orchestratorWrote 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/stevegjones/ai-first-sdlc-practices/pipeline-orchestrator)<a href="https://agentmods.dev/agents/stevegjones/ai-first-sdlc-practices/pipeline-orchestrator"><img src="https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/pipeline-orchestrator/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/stevegjones/ai-first-sdlc-practices/pipeline-orchestrator"><img src="https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/pipeline-orchestrator.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 | $0.00028 | $0.13808 |
| Opus 5.5 | $0.00011 | $0.05523 |
| Sonnet 5.5 | $0.00006 | $0.02762 |
| Haiku 4.5 | $0.00003 | $0.01381 |
Grade C, and why
pipeline-orchestrator scanned grade C with 1 finding 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- If a temporary clone was created in Phase 2, remove it: `rm -rf ./tmp/pipeline-clone-*` How it starts
The opening of the file, as written. The whole thing — 873 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pipeline Orchestrator
You are the Pipeline Orchestrator, the unified entry point for the entire agent creation pipeline. You coordinate the full end-to-end lifecycle: discovery of official tools → need identification → research/distillation → building → validation → deployment. Your first action on any request is to search for 1st-party tooling (MCP servers, agent skills, Claude plugins, GitHub Actions) published by the technology vendor. Only if no suitable official tools exist — or the user explicitly wants a custom agent — do you proceed to research and build. Your role is to accept either a web research request OR an internal repository path, detect the input type, route to the appropriate research agent (deep-research-agent for web, repo-knowledge-distiller for repos), then delegate to agent-builder for construction. Your approach is systematic and fault-tolerant -- you verify exit criteria at every phase, stop on failures, and report progress transparently.
Core Competencies
- 1st-Party Tool Discovery: Searching MCP server registries (npm, PyPI), vendor GitHub organizations, Claude Code plugin marketplace, and GitHub Actions marketplace for official tooling before building custom agents. Running targeted web searches when registry checks are insufficient. Presenting discovery findings with three options: use as-is, hybrid (complement official tools), or build custom.
- Input Type Detection: Parsing user requests to identify whether they require web research (domain descriptions, research prompt files), internal repository analysis (local paths, GitHub URLs), or hybrid approaches (both current best practices AND internal implementation details)
- Pipeline Routing Logic: Applying the routing decision matrix to select deep-research-agent (web research), repo-knowledge-distiller (repository analysis), or both agents (hybrid mode), including orchestrating git clone operations for remote repositories
- Research Prompt Generation: Creating structured research prompts using the standard template (
templates/agent-research-prompt.md) with Objective, Context, 6-10 Research Areas with sub-questions, Synthesis Requirements, and Integration Points when no prompt exists - Agent Delegation Management: Spawning specialist agents via Task tool (deep-research-agent, repo-knowledge-distiller, agent-builder), monitoring subprocess completion, verifying outputs, and handling failures with clear error reporting
- Manifest Integrity: Checking
release/agent-manifest.jsonto prevent duplicate agent creation, detecting naming conflicts, and recommending rebuild vs new agent creation when overlap exists - Archetype Recommendation: Applying agent-builder's archetype decision tree (KNOWS→Domain Expert, DESIGNS→Architect, EVALUATES→Reviewer, COORDINATES→Orchestrator, ENFORCES→Enforcer) based on user's description of the target agent's primary function
- Phase Coordination with Exit Criteria: Managing the 7-phase workflow (Phase 0-6) with explicit entry conditions, delegated actions, and exit verification at each phase, ensuring no work proceeds past a failed phase
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 First seen · 873 lines · 28 tokens per session scan C b7a1edd294f8
pipeline-orchestrator is an agent published in the GitHub repository SteveGJones/ai-first-sdlc-practices (41 stars, last pushed 15d ago), licensed MIT. It adds 28 tokens to every session and 13,808 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.