report

Invoke the analyst agent to run all three operations and deliver the integrated insight report.

Workflow

1. Confirm data exists

Application data uses the split layout defined in references/tracker-schema.md — read it before any read or write.

Read {user_dir}/CareerNavigator/tracker.json, {user_dir}/CareerNavigator/ExperienceLibrary.json, {user_dir}/CareerNavigator/profile.md, and {user_dir}/CareerNavigator/artifacts-index.json. If the ExperienceLibrary is empty:

"Your ExperienceLibrary is empty. Run /career-navigator:add-source to add a resume first."

If the tracker has no applications, note this inline — pattern analysis will be limited but the other two operations can still run.

tracker.json now holds summary rows only. Operations 2 and 3 (transferable strengths, AI displacement) run off those rows plus the ExperienceLibrary and need nothing else. Operations 1 and 4 (outcome patterns, market benchmark) reason over stage transitions and note text, which live in the per-application detail files — so before the handoff, collect {user_dir}/CareerNavigator/ + detail_file for every summary row with stage_count greater than 0 or outcome != "pending", and skip the rest. Do not load contacts files; no analyst operation uses them.

1.5 Ensure ATS scores exist (resume artifacts)

Read {user_dir}/CareerNavigator/artifacts-index.json.

If any artifact with type: "resume" is missing ats_score (or has ats_score: null):

• Invoke resume-score for that resume artifact (reference its filename so the skill can look it up and run the ATS check pass).
• If keyword coverage can’t be computed due to missing JD context, still proceed with formatting/narrative ATS checks and write whatever ats_score the ATS pass can compute.
• Re-check {user_dir}/CareerNavigator/artifacts-index.json after the ATS pass(es). If scores are still missing, note the remaining ATS gaps in the report highlights.