Build a dev-workflow generator with test-driven agent coding
The author built an open-source system that generates project-specific dev workflows from an interview, entirely with Claude Code, which wrote and debugged 550+ tests. Task: Suggested implementation: Use Claude Code to run a /bootstrap interview that generates project-specific slash commands, guard hooks, pre-push checks, and agent configs. The agent writes main code files and executes a test-driven workflow to produce and debug 550+ tests. Human role involves defining the initial stack and reviewing generated configurations.; tools: Claude Code. Human role: Source evidence is unavailable regarding specific human roles or manual interventions beyond the initial interview setup.; limitations: Prerequisites include installing Claude Code CLI and Backlog.md CLI. Integration uncertainty exists regarding compatibility with non-standard web stacks. Source reliability is low due to inaccessible original content. Human approval is required before deploying generated hooks to production repositories..
Tools
Published on Agent Work Atlas:
Overview
The author built an open-source system that generates project-specific dev workflows from an interview, entirely with Claude Code, which wrote and debugged 550+ tests.
Workflow
Suggested implementation: Use Claude Code to run a /bootstrap interview that generates project-specific slash commands, guard hooks, pre-push checks, and agent configs. The agent writes main code files and executes a test-driven workflow to produce and debug 550+ tests. Human role involves defining the initial stack and reviewing generated configurations.
Reported result
Actual outcomes are unknown as the source returned a 403 error and no execution logs are available. Proposed acceptance check: Verify that the generated workflow correctly identifies the tech stack and passes all 550+ unit tests without manual intervention.
Human role
Source evidence is unavailable regarding specific human roles or manual interventions beyond the initial interview setup.
Limitations
Prerequisites include installing Claude Code CLI and Backlog.md CLI. Integration uncertainty exists regarding compatibility with non-standard web stacks. Source reliability is low due to inaccessible original content. Human approval is required before deploying generated hooks to production repositories.
Try it
Act as a DevOps engineer. I want to set up an agentic workflow for my project [replaceable: project name] using Claude Code. First, install Claude Code CLI and Backlog.md CLI. Run the /bootstrap command and answer the interview questions about my stack: [replaceable: tech stack details]. Generate the project-specific slash commands, guard hooks, and pre-push checks. Then, instruct the agent to write the main code files and create 550+ tests in a test-driven manner. Run self-audits with four parallel reviews. Before publishing, perform a dry run on a sample module to verify test coverage. Output the configuration files to [replaceable: output directory]. If any tests fail, debug them iteratively. Do not push to the main branch until I confirm the audit results. Ask me for clarification if the stack details are ambiguous.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.