Workflow Integrator

You've designed repeatable AI-assisted workflows. The AI is a designed tool within your research process, not a chat partner.

If this is you

You use Claude Code for file-heavy work — processing corpora, maintaining structured notes, cross-referencing sources
You've designed repeatable workflows with defined inputs, outputs, and verification steps
You think explicitly about when not to use AI as much as when to use it

Your risk

⚠ False systematicity

Repeatable workflows create an impression of rigour that the underlying AI outputs may not deserve. The interpretive judgments embedded in your prompts are invisible once they're in a template. Errors replicate at scale rather than staying isolated — by the time they surface, they're woven through the whole output.

To reach Level 4

Workflow Integrator → Agent Orchestrator

Conceptual Shift

"The human role shifts from doing to specifying and reviewing."

At Level 4, you define what needs to happen clearly enough to delegate; you review whether what came back is what was wanted; you intervene when the agent goes wrong. The skill is no longer execution — it is specification and judgement. Automation also introduces error modes that are qualitatively different: an error in step 3 of a five-step automated chain may only surface in the final output, plausibly and confidently.

Experiment

Write a two-step script: something where the output of Step 1 feeds Step 2 — for example, generating structured summaries from a document set and then synthesising them across documents. Keep the scope small enough that you can verify every intermediate output. Run it. Read the intermediate outputs before running Step 2. The goal is to experience what "the chain completed but something went wrong at step 3" feels like before you are relying on a larger process.

Practice

Build verification gates into every automated chain. At minimum: review a sample of intermediate outputs before the next step runs.
Add steps incrementally, not all at once. Each added step multiplies the potential for error propagation.
Respect the threshold: this level genuinely requires programming skill, or a close working relationship with someone who has it.

Note. Consider whether you have a specific use case that genuinely requires Level 4 before moving forward. Level 3 is the appropriate endpoint for most researchers who want to work systematically with files.

Next level →

Agent Orchestrator

← Level 2: Context-Aware Collaborator