$npx -y skills add Egonex-AI/Understand-Anything --skill understand-diffUse when you need to analyze git diffs or pull requests to understand what changed, affected components, and risks
| 1 | # /understand-diff |
| 2 | |
| 3 | Analyze the current code changes against the knowledge graph in the project's data directory (`.ua/knowledge-graph.json`, or the legacy `.understand-anything/knowledge-graph.json` when that directory is present). |
| 4 | |
| 5 | ## Graph Structure Reference |
| 6 | |
| 7 | The knowledge graph JSON has this structure: |
| 8 | - `project` — {name, description, languages, frameworks, analyzedAt, gitCommitHash} |
| 9 | - `nodes[]` — each has {id, type, name, filePath?, summary, tags[], complexity, languageNotes?} |
| 10 | - Code node types: file, function, class, module, concept |
| 11 | - Non-code node types: config, document, service, table, endpoint, pipeline, schema, resource |
| 12 | - Domain/knowledge node types: domain, flow, step, article, entity, topic, claim, source |
| 13 | - IDs use the node type as prefix, e.g. `file:path`, `function:path:name`, `config:path`, `article:path` |
| 14 | - `edges[]` — each has {source, target, type, direction, weight} |
| 15 | - Key types: imports, contains, calls, depends_on, configures, documents, deploys, triggers, contains_flow, flow_step, related, cites |
| 16 | - `layers[]` — each has {id, name, description, nodeIds[]} |
| 17 | - `tour[]` — each has {order, title, description, nodeIds[]} |
| 18 | |
| 19 | ## How to Read Efficiently |
| 20 | |
| 21 | 1. Use Grep to search within the JSON for relevant entries BEFORE reading the full file |
| 22 | 2. Only read sections you need — don't dump the entire graph into context |
| 23 | 3. Node names and summaries are the most useful fields for understanding |
| 24 | 4. Edges tell you how components connect — follow imports and calls for dependency chains |
| 25 | |
| 26 | ## Instructions |
| 27 | |
| 28 | 1. **Resolve the data directory `$UA_DIR`.** Run `UA_DIR=$([ -d .understand-anything ] && echo .understand-anything || echo .ua)` — this is the legacy `.understand-anything/` when it already exists, otherwise the new `.ua/`. Check that `$UA_DIR/knowledge-graph.json` exists. If not, tell the user to run `/understand` first. |
| 29 | |
| 30 | 2. **Get the changed files list** (do NOT read the graph yet): |
| 31 | - If on a branch with uncommitted changes: `git diff --name-only` |
| 32 | - If on a feature branch: `git diff main...HEAD --name-only` (or the base branch) |
| 33 | - If the user specifies a PR number: get the diff from that PR |
| 34 | |
| 35 | 3. **Read project metadata and check graph freshness** — use Grep or Read with a line limit to extract the `"project"` section, including `gitCommitHash` as `GRAPH_COMMIT_RAW`, then: |
| 36 | - Resolve it as a commit before using it in any Git diff. From the project root, compare the resolved commit with `git rev-parse HEAD` and inspect project-scoped committed and working-tree changes: |
| 37 | ```bash |
| 38 | GRAPH_COMMIT=$(git rev-parse --verify --end-of-options "${GRAPH_COMMIT_RAW}^{commit}" 2>/dev/null) |
| 39 | git rev-parse HEAD |
| 40 | git diff --name-only "$GRAPH_COMMIT" HEAD -- . |
| 41 | git diff --cached --name-only -- . |
| 42 | git diff --name-only -- . |
| 43 | git ls-files --others --exclude-standard -- . |
| 44 | ``` |
| 45 | - The `-- .` pathspec is required: commits that only touch a sibling monorepo project must not make this graph stale. A hash mismatch alone is not stale when the project diff is empty. |
| 46 | - Ignore the selected data directory (`.ua/` or legacy `.understand-anything/`) in every command's output because it contains generated graph artifacts, not project source drift. |
| 47 | - If the committed diff or any working-tree command reports project files, warn before impact analysis that the graph may omit those changes. Suggest: Run `/understand` to refresh the graph. |
| 48 | - Run the commit diff only when `GRAPH_COMMIT_RAW` resolves successfully. If the graph commit or Git metadata is missing, invalid, or unavailable, give a brief best-effort warning and continue instead of blocking. |
| 49 | |
| 50 | 4. **Find nodes for changed files** — for each changed file path, use Grep to search the knowledge graph for: |
| 51 | - Nodes with matching `"filePath"` values (e.g., `grep "changed/file/path"`) |
| 52 | - This finds file-level nodes (including non-code types) AND function/class nodes defined in those files |
| 53 | - Note the `id` values of all matched nodes |
| 54 | |
| 55 | 5. **Find connected edges (1-hop)** — for each matched node ID, Grep for that ID in the edges to find: |
| 56 | - What imports or depends on the changed nodes (upstream callers) |
| 57 | - What the changed nodes import or call (downstream dependencies) |
| 58 | - These are the "affected components" — things that might break or need updating |
| 59 | |
| 60 | 6. **Identify affected layers** — Grep for the matched node IDs in the `"layers"` section to determine which architectural layers are touched. |
| 61 | |
| 62 | 7. **Provide structured analysis**: |
| 63 | - **Changed Components**: What was directly modified (with summaries from matched nodes) |
| 64 | - **Affected Components**: What might be impacted (from 1-hop edges) |
| 65 | - **Affected Layers**: Which architectural layers are touched and cross-layer concerns |
| 66 | - **Risk Assessment**: Based on node `complexity` values, number of cross-layer edges, and blast |