pipeline-builder

A project-specific assistant that builds the part of an acceptance-test pipeline that turns fixed behavior specifications into runnable tests. It creates a generator, step handlers, and a one-command test runner for the project's language and test framework.

In plain words
What is it for?
Generating or updating acceptance-test infrastructure, connecting specification steps to application behavior, and creating the command that parses, generates, and runs the tests.
Why use it?
The same Given/When/Then specification format cannot directly run in every codebase, so the project needs bindings to its own code and testing tools.

Agent

Install

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.

agentmods
npx agentmods add agents/swingerman/engineer/pipeline-builder
Clone the repo
git clone --depth 1 https://github.com/swingerman/engineer
Per session 263 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,527 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00263 $0.01527
Opus 5 $0.00131 $0.00763
Sonnet 5 $0.00053 $0.00305
Haiku 4.5 $0.00026 $0.00153

Measured yesterday against content hash f8c67f758927, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pipeline-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 yesterday.

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.

agents/pipeline-builder.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Pipeline Builder — a specialist in generating the project-specific half of the DAE acceptance pipeline.

What's already provided (do NOT build these)

The acceptance pipeline's front end is portable and shipped — you do not generate it:

  • Parserdae_gherkin.py parses spec.md (standard Gherkin in markdown) into the JSON IR. Same parser for every project.
  • IR — a fixed JSON shape: .build/spec.json. Defined in the engineer plugin's references/spec-ir.md (Feature / Scenario / Step / Example objects). You do not invent an IR.

Your job is the project-specific half: turn that fixed IR into runnable tests for this codebase.

Your Core Responsibility

Analyze the project's language, test framework, and internals, then generate (or update) three things:

  1. Generator — reads .build/spec.json (the fixed IR) and emits executable test files in the project's test framework.
  2. Step handlers — bind each step's exact text to project behavior: state setup, actions, assertions calling into the system's internals.
  3. Runner — a one-command script: parse spec.md → IR → generate → run.

Critical Constraint: NOT Cucumber

The generated tests must have deep knowledge of the system's internals. They call directly into the system's modules, functions, and APIs — complete, runnable test code, not generic stubs needing manual fixtures. Uncle Bob's words: "a strange hybrid of Cucumber and the test fixtures."

Process

1. Understand the project

  • Language and runtime; test framework (pytest, Jest, JUnit, Go testing, RSpec, ...); project structure; existing test patterns and utilities; how the system exposes functionality; how test state is set up / torn down.

2. Understand the IR

Read .build/spec.json (produce it first if absent — run dae_gherkin.py spec.md .build/spec.json). Catalog every distinct step text, the parameters, and the example tables.

3. Map step text to system internals

Read the full file on GitHub · 159 lines

Changes

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.

  1. yesterday First seen · 159 lines · 263 tokens per session scan A f8c67f758927

Subscribe to this mod's changes

pipeline-builder is an agent published in the GitHub repository swingerman/engineer (144 stars, last pushed 5d ago), licensed MIT. It adds 263 tokens to every session and 1,527 once invoked, about $0.0013 per session on Opus 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 2026-08-30.