graphify-query

graphify-query: BFS/DFS over the code graph

Wraps python -m graphify query for an existing project graph. Returns nodes ranked by relevance to the query terms, with source locations and community membership.

When to invoke

• The project has graphify-out/graph.json (built by /graphify-ingest or /graphify-update).
• The user asks an open-ended code-structural question: "what's connected to auth?", "explore the OAuth area", "what touches getOAuthConfig?".
• The question is not a single-node lookup (that's graphify-explain) and not a two-node trace (that's graphify-path).

If graph.json doesn't exist at the resolved target, tell the user /graphify-ingest is needed first. Don't invent.

Steps

Step 1 — Resolve target path

TARGET="${1:-$PWD}"
TARGET=$(cd "$TARGET" && pwd)
GRAPH="$TARGET/graphify-out/graph.json"
[ -f "$GRAPH" ] || { echo "No code graph at $GRAPH. Run /graphify-ingest first."; exit 1; }

Step 2 — Resolve a working Python with graphifyy

PYTHON=""
PIN_FILE="$TARGET/graphify-out/.graphify_python"
if [ -f "$PIN_FILE" ]; then
    PINNED=$(cat "$PIN_FILE")
    if [ -x "$PINNED" ] && "$PINNED" -c "import graphify" 2>/dev/null; then
        PYTHON="$PINNED"
    fi
fi
if [ -z "$PYTHON" ]; then
    for cand in python3 python; do
        if command -v "$cand" >/dev/null 2>&1 && "$cand" -c "import graphify" 2>/dev/null; then
            PYTHON="$cand"; break
        fi
    done
fi
[ -z "$PYTHON" ] && { echo "graphifyy not installed. Run bin/setup-graphify.sh on this project."; exit 1; }

Step 3 — Run the query

cd "$TARGET" && "$PYTHON" -m graphify query "$QUERY" [--dfs] [--budget N]

Defaults:

• BFS traversal (drop --dfs unless the user asked for depth-first).
• Token budget 1500 for normal questions; bump to 3000 for "tell me everything about X" wide queries.
• Output goes to stdout in plain text already formatted for LLM consumption.

Step 4 — Synthesize and cite

graphify's output lists nodes in relevance order with src= pointers. Read it as the raw retrieval, then write a short answer in chat:

• Name the start nodes the BFS found (these tell the user what their keywords matched).
• Group results by community when ≥3 nodes share one — name the community using the labels in graphify-out/labels.json so users see "OAuth Config and Expo Constants" not "community 3".
• Cite source files with full paths (graphify already includes them).

If the user asks a follow-up about one node, switch to /graphify-explain "<node>". If they ask "how does A reach B", switch to /graphify-path.

Tips for query phrasing

graphify matches start nodes by keyword substring against node labels, not natural language. "What calls X?" usually returns "no matching nodes" because "calls" isn't in any label. Strip filler — query just X or oauth refresh. The BFS does the rest.

Cost

$0. Pure Python over the existing JSON graph. No LLM tokens consumed unless the user asks for synthesis afterward.

Cross-skill notes

• The wiki-query skill already routes some code-structural questions to graphify CLI. Use this skill when the user explicitly invokes /graphify-query or names the graph; let wiki-query handle ambient "what is X" type questions that may or may not map to the graph.
• graphify-out/memory/ (when present) is graphify's Q&A feedback store. Optional follow-up: python -m graphify save-result --question Q --answer A --type query to record the exchange.