dx-plan-resolve

You resolve risks and issues found during plan validation by researching the codebase for concrete solutions, then updating implement.md with specific fix instructions.

Use ultrathink for this skill — solving risks requires deep reasoning about patterns and codebase specifics.

Output

You run in a forked context. Before emitting any chat output, determine whether you were invoked by the orchestrator (dx-agent-all) or standalone — see plugins/dx-core/shared/orchestration-check.md:

ORCHESTRATED=0
FLAG=".ai/run-context/orchestrating.flag"
if [ -f "$FLAG" ]; then
  AGE=$(( $(date +%s) - $(date -r "$FLAG" +%s) ))
  [ "$AGE" -lt 7200 ] && ORCHESTRATED=1
fi

• If $ORCHESTRATED == 1 (orchestrator path): write the full report to $SPEC_DIR/resolve-report.md and emit ONLY the ## Return block to chat.
• If $ORCHESTRATED == 0 (standalone path): write the same report AND emit the human-friendly summary marked <!-- standalone-only --> below, followed by the ## Return block at the very end.

Per-phase progress lines during the run are allowed in both paths.

This skill writes its full report (per-issue resolution actions and updated implement.md diff summary) to $SPEC_DIR/resolve-report.md.

1. Locate the Spec Directory

SPEC_DIR=$(bash .ai/lib/dx-common.sh find-spec-dir $ARGUMENTS)

Read from $SPEC_DIR:

• implement.md (required)
• research.md (required — for codebase context)
• explain.md (for requirement context)

2. Identify Issues to Resolve

Scan implement.md for:

1. Risks section — anything listed under ## Risks
2. Blocked steps — steps with **Status:** blocked
3. Vague instructions — steps that say "may need", "might require", "consider", "TBD" without a concrete solution

Collect each issue as a task to resolve.

If no issues found → print No risks or issues to resolve — plan is ready. and STOP.

3. Research Solutions

For each issue, search the codebase for how the project already solves similar problems.