From AI Conversation to Reusable Workflow on Work You Can Explain

Don’t Let a Useful AI Conversation Disappear

Fri, 24 Jul 2026 00:00:00 -0500

A useful AI conversation is not durable documentation. It may contain false starts, hidden context, private information, and conclusions that only make sense inside the exchange. Capture the verified lesson in your own record.

Record the problem

State the original objective, starting state, constraints, and exact failure or question. Include software versions, current directory, input shape, and error text only when relevant and safe.

Do not begin with “AI fixed it.” A future reader needs the technical problem.

The Capture–Verify–Crystallize Loop

Fri, 24 Jul 2026 00:00:00 -0500

An AI-assisted result becomes reusable only after it works without the conversation. Use a three-stage loop: capture the result and context, verify the claims and behavior, then crystallize the confirmed lesson into the smallest useful asset.

Capture enough context

Begin with the learning record:

objective and starting state;
constraints and policy boundary;
attempted steps;
result that appeared successful;
material AI assistance;
risks and assumptions; and
evidence still needed.

Capture is not endorsement. It prevents important context from disappearing before review.

Should This Be a Note, Checklist, Script, or Skill?

Fri, 24 Jul 2026 00:00:00 -0500

Choose the smallest reusable form that preserves the verified value. Use a note for understanding, a checklist for human judgment, a script for deterministic operations, a project convention for repository-wide behavior, a skill for agentic interpretation, and automation only when a stable trigger should run a known action.

Note: preserve understanding

Use a note when the main value is an explanation, decision, example, or source. A note can answer why a configuration exists or what an error taught you.

Write a Bounded Task for an AI Coding Agent

Thu, 03 Sep 2026 00:00:00 -0500

A useful AI coding-agent task names the outcome, relevant starting context, exact targets, constraints, authority, success criteria, and evidence to report. It also states what should remain unchanged and what ambiguity should stop the work.

The task does not need to prescribe every keystroke. It needs a boundary strong enough for you to decide whether the result belongs.

Why vague requests expand risk

Consider this request:

Clean up this project and improve the README.

The agent must infer what “clean up” means. It might reformat code, update dependencies, delete files, rewrite documentation, change repository settings, or make a commit. Even a technically competent result may solve a different problem than the one the student intended.

Automate Deterministic Work with Code

Fri, 24 Jul 2026 00:00:00 -0500

Use code instead of repeated AI prompts when the same explicit inputs should produce defined outputs through known operations. Keep AI at the interpretation boundary: it can help explain an unfamiliar case or draft code, but deterministic execution should be inspectable, tested, and repeatable.

Define the contract

Before scripting, write:

Inputs:
Outputs:
Files or systems changed:
Valid and invalid conditions:
Expected failures:
Verification:
Rollback:

For example, renaming assignment files by a documented convention can be deterministic. Deciding whether a student’s reflection demonstrates understanding requires judgment and should not be hidden in a rename script.

Use `AGENTS.md` to Teach an AI About a Repository

Fri, 24 Jul 2026 00:00:00 -0500

Use AGENTS.md for durable guidance that should apply across many agent tasks in a repository. Put project conventions, reliable setup and verification commands, scope boundaries, and review expectations there. Keep the current objective, temporary context, and one-time preferences in the task prompt.

OpenAI’s current AGENTS.md documentation explains that Codex reads instruction files before working. Instructions closer to the current directory take precedence over broader repository guidance, and an AGENTS.override.md file can replace an AGENTS.md file at the same level. Confirm the files Codex actually loaded before relying on those rules.

Package a Repeatable AI Workflow as a Skill

Fri, 24 Jul 2026 00:00:00 -0500

Package a workflow as a skill when it recurs across tasks, contains several steps, requires some agent judgment, and benefits from shared instructions, references, scripts, templates, and a consistent output. Use ordinary code for deterministic operations and AGENTS.md for durable repository rules.

OpenAI’s current Codex Build skills documentation defines an agent skill as a directory with a required SKILL.md. That file supplies a name, a description that helps ChatGPT or Codex decide when to use the skill, and the workflow instructions. The directory can also include scripts, references, and assets. Follow the validation tools in the Codex environment where the skill will run; other products may package or distribute the same workflow differently.

Adapt Shared Agent Skills to Your Own Workflow

Mon, 24 Aug 2026 00:00:00 -0500

An existing agent skill is a starting point, not an authority. Before you install, combine, or adapt one into a skill you already maintain, inspect what it assumes, compare it with your current workflow, and test the meaningful difference on a small safe task.

Keep at most one bounded adaptation at a time. Preserve the source, version, license, rationale, rejected alternatives, and rollback path. A passing example is evidence about that task and environment. It is not proof that one skill is universally better.

Grow With Skills: Improve the Agent While You Work

Tue, 28 Jul 2026 00:00:00 -0500

A completed task can reveal how a skill should improve, but the task is evidence, not proof. Record the observed gap, separate the reusable lesson from the task’s local context, and propose one bounded change. A shared skill can supply another design to compare, but it is a source, not an upstream workflow you must follow. Keep a local revision only when it addresses an observed need, preserves attribution and boundaries, and passes fresh-task tests.

Automation Is the Last Step, Not the First

Fri, 24 Jul 2026 00:00:00 -0500

Automate a workflow only after you can perform it manually, explain its decisions, and verify its result. Begin with a working procedure. Document it, move deterministic steps into code, test the code, and add scheduling only when monitoring and recovery are ready.

Automation multiplies the workflow you give it. If the workflow is unclear, automation repeats unclear decisions faster.

Perform the work manually first

Complete the task on a small, representative example. Record: