df:research-objective

npx skills add majiayu000/claude-skill-registry --skill df-research-objective --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

<objective> Research how to implement an objective. Spawns df-objective-researcher agent with objective context. **Note:** This is a standalone research command. For most workflows, use `/df:plan-objective` which integrates research automatically. **Use this command when:** - You want to research without planning yet - You want to re-research after planning is complete - You need to investigate before deciding if an objective is feasible **Orchestrator role:** Parse objective, validate against roadmap, check existing research, gather context, spawn researcher agent, present results. **Why subagent:** Research burns context fast (WebSearch, Context7 queries, source verification). Fresh 200k context for investigation. Main context stays lean for user interaction. </objective> <context> Objective number: $ARGUMENTS (required) Normalize objective input in step 1 before any directory lookups. </context> <process> ## 0. Initialize Context ```bash INIT=$(node ~/.claude/devflow/bin/df-tools.cjs init objective-op "$ARGUMENTS") ``` Extract from init JSON: `objective_dir`, `objective_number`, `objective_name`, `objective_found`, `commit_docs`, `has_research`. Resolve researcher model: ```bash

ls .planning/objectives/${OBJECTIVE}-*/RESEARCH.md 2>/dev/null
Objective section already loaded in OBJECTIVE_INFO
echo "$OBJECTIVE_INFO" | jq -r '.section'
cat .planning/REQUIREMENTS.md 2>/dev/null
cat .planning/objectives/${OBJECTIVE}-*/*-CONTEXT.md 2>/dev/null
grep -A30 "### Decisions Made" .planning/STATE.md 2>/dev/null