https://vibecodetextbook.com/ Field notes on coding with AI agents: harness walkthroughs, repeatable workflows, the repos worth reading, and honest measurements of what the tools actually do. 2026-09-10T00:00:00Z https://vibecodetextbook.com/posts/prompt-patterns-coding-agents-plan-mode-skills.html https://vibecodetextbook.com/posts/prompt-patterns-coding-agents-plan-mode-skills.html Four prompt patterns from the Claude Code, Codex, and Gemini CLI docs, with exact commands and file formats, plus a script that writes one saved prompt three ways. 2026-09-10T00:00:00Z https://vibecodetextbook.com/posts/coding-agent-security-prompt-injection-mcp-secrets.html https://vibecodetextbook.com/posts/coding-agent-security-prompt-injection-mcp-secrets.html Prompt injection via tool results, MCP server trust, and secrets in config: what the Claude Code, Codex, and Gemini CLI docs commit to, plus an auditor to run. 2026-09-07T00:00:00Z https://vibecodetextbook.com/posts/when-to-stop-a-coding-agent.html https://vibecodetextbook.com/posts/when-to-stop-a-coding-agent.html The budget flags harnesses ship, the loop signatures that predict a failed run, what the research says about early stopping, and a transcript analyser. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/what-one-coding-task-costs-tokens-dollars-minutes.html https://vibecodetextbook.com/posts/what-one-coding-task-costs-tokens-dollars-minutes.html A method for costing an agent task across an interactive session, a headless run, and a cheaper model, with a price calculator and a dataset of published numbers. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/what-is-a-coding-agent-harness.html https://vibecodetextbook.com/posts/what-is-a-coding-agent-harness.html A coding-agent harness is the loop around the model. Here is what each part does, read from three real harnesses, with a 100-line runnable version you can inspect. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/time-to-green-ten-small-tasks-method.html https://vibecodetextbook.com/posts/time-to-green-ten-small-tasks-method.html A method and a first dataset for timing a coding agent from prompt to first passing check: ten defined tasks, a runner script, and only honestly filled columns. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/tests-first-with-an-agent-red-green.html https://vibecodetextbook.com/posts/tests-first-with-an-agent-red-green.html The red-green loop for agents: why a test that never failed proves nothing, hook-based gates, the lucky-pass numbers, and a script that enforces the order. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/swe-bench-explained-what-the-numbers-measure.html https://vibecodetextbook.com/posts/swe-bench-explained-what-the-numbers-measure.html How a SWE-bench score is produced, what Verified changed, and five papers on harness effects, lucky passes, contamination, and realistic prompts, with a dataset. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/swe-agent-agent-computer-interface-explained.html https://vibecodetextbook.com/posts/swe-agent-agent-computer-interface-explained.html What SWE-agent's ACI paper argued in 2024, what the repo's tools directory looks like at commit 3ea751c0, and why its own authors now recommend a 100-line successor. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/spec-first-prompting-coding-agent.html https://vibecodetextbook.com/posts/spec-first-prompting-coding-agent.html A six-section spec template for coding agents, a worked CLI example, the published cost of underspecified prompts, and a linter to run before handing over. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/reviewing-agent-diffs-checklist.html https://vibecodetextbook.com/posts/reviewing-agent-diffs-checklist.html Ten patterns behind most damage in agent diffs, why a second model makes a good first reviewer, and a scanner that flags them with line references. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/refactor-with-an-agent-characterization-tests.html https://vibecodetextbook.com/posts/refactor-with-an-agent-characterization-tests.html Restructuring a legacy script with a coding agent: the same characterization tests run against old and new code, one refactor per turn, fixes in their own commit. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/openhands-architecture-explained.html https://vibecodetextbook.com/posts/openhands-architecture-explained.html How OpenHands is built at v1.16.0. The agent step loop, workspaces and sandboxes, the condenser, the confirmation gate, and what the README now calls Agent Canvas. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/omnigent-meta-harness-explained.html https://vibecodetextbook.com/posts/omnigent-meta-harness-explained.html How I picked this month's fastest-rising agent repo with a GitHub search, what Omnigent's README says it is, its policy and sandbox model, and what is unverified. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/mcp-server-tutorial-python.html https://vibecodetextbook.com/posts/mcp-server-tutorial-python.html What the Model Context Protocol changes for coding agents, the JSON-RPC on the wire, and a dependency-free Python stdio server exercised with a scripted client. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/headless-coding-agent-ci.html https://vibecodetextbook.com/posts/headless-coding-agent-ci.html The flags for non-interactive Claude Code and Codex CLI runs, the caps that bound a job, the risks the docs name, and a CI wrapper exercised in dry-run mode. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/git-worktrees-branches-agent-sessions.html https://vibecodetextbook.com/posts/git-worktrees-branches-agent-sessions.html A worktree and branch per agent task, commits after every green, deny rules for destructive git, and a helper script exercised on a real repository. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/context-files-claude-md-agents-md-cursor-rules.html https://vibecodetextbook.com/posts/context-files-claude-md-agents-md-cursor-rules.html Where each coding agent looks for standing instructions, how the files load and merge, the size limits, and an audit script that finds every context file in a repo. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/coding-agent-permission-models-compared.html https://vibecodetextbook.com/posts/coding-agent-permission-models-compared.html The permission models of Claude Code, Codex CLI, Gemini CLI, the OpenHands SDK, and Omnigent, read from their docs on 2026-09-05, with a working settings file. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/build-a-cli-tool-with-an-agent-from-a-spec.html https://vibecodetextbook.com/posts/build-a-cli-tool-with-an-agent-from-a-spec.html A method walkthrough for building a small CSV CLI with a coding agent: the spec, the prompts in order, the acceptance tests, and the checks that say it is done. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/aider-repo-map-edit-formats-explained.html https://vibecodetextbook.com/posts/aider-repo-map-edit-formats-explained.html Aider read at commit 5dc9490b. How the tree-sitter repo map is ranked and budgeted, what each edit format does, and a stdlib repo-map script you can run. 2026-09-05T00:00:00Z https://vibecodetextbook.com/posts/add-tests-to-a-legacy-script-with-an-agent.html https://vibecodetextbook.com/posts/add-tests-to-a-legacy-script-with-an-agent.html A step-by-step method for getting characterization tests onto an untested script with a coding agent: survey, seams, pin the actual behaviour, then talk about bugs. 2026-09-05T00:00:00Z