Reduce cognitive load in Step 16 data-source workshop content

[Copilot]

Step 16 introduced too many new concepts at once by explaining output plumbing, expression syntax, heredoc mechanics, and data-fetch commands inline. This update trims that overload while keeping the hands-on workflow edits intact and making the explanation easier to scan.

• Tighten the conceptual intro

  • Reworked the “Understand the data-flow pattern” section into short causal prose: fetch data first, then pass saved results into the prompt.
  • Removed the inline mini-taxonomy of shell/output/expression concepts that was driving concept count up.

• Offload secondary mechanics to existing side quests

  • Replaced inline explanations of $GITHUB_OUTPUT, here-document syntax, and scripted-vs-agentic decisions with a compact TIP that links to the existing side quests.
  • Kept the main step focused on the learner’s immediate task instead of teaching multiple subtopics in parallel.

• Preserve the active-learning path

  • Kept all five code blocks and the same workflow-editing flow.
  • Retained the prediction/try-it prompts, but simplified surrounding prose and checkpoint wording so the learner stays anchored on the data-flow pattern.

• Keep technical identifiers precise where they matter

  • Preserved exact workflow field names and step IDs where learners need to copy or recognize them (steps:, id: recent, id: issues, commit_log, permissions:, network.allowed-domains).

Example of the updated pattern framing:

[gh-aw workflows](https://github.github.com/gh-aw/introduction/overview/) run inside GitHub Actions, so your workflow can fetch live repository data before the AI writes anything. In this step, you will use shell steps to collect data and a later prompt section to turn that data into a summary.

Think of it as a handoff. First, the workflow gathers facts in a predictable way. Then, the prompt reads those saved results and asks the AI to explain what matters.