← ClaudeAtlas

enrichlisted

Run the file-analyser subagent against one or more substrate nodes to produce per-node semantic enrichment sidecars. Resolves a path / node-id / batch spec to substrate nodes; spawns file-analyser per node; validates the resulting sidecar against the schema; updates the manifest's `enrichment` block.
opendatadiscovery/odd-team · ★ 0 · AI & Automation · score 75
Install: claude install-skill opendatadiscovery/odd-team
# Enrich an ontology node (DOC-164 slice 5+) Drive the agentic-code-ontology pipeline for one or more substrate nodes. Each invocation produces (or refreshes) Markdown sidecars at `lineage/{repo}/understanding/{slug}.md`, written by the `file-analyser` subagent with the live-URL-only doc rule and Gate 9 code-anchor mandate. This skill is the maintainer-facing entry point for the per-node enrichment layer per `adrs/drafts/agentic-code-ontology.md` revision 2. ## Prerequisite The substrate scaffold (DOC-164 slices 1-4) must have run against the target repo. `lineage/{repo}/nodes.jsonl` must exist. If it doesn't, run `/scan scanners/code-lineage/{axis}.md` first or invoke the substrate's `lineage-extractor scan {repo} --full`. ## Argument forms | Form | Behaviour | |---|---| | `/enrich <path>` | Single-file enrichment. Resolves `<path>` (relative to a repo) to its substrate node(s) and enriches each. If the path matches multiple nodes (e.g. a controller file with multiple methods), enrich the file-level node first, then the maintainer can drill into method nodes via `/enrich --node <id>`. | | `/enrich --node <node-id>` | Single-node enrichment by node ID (verbatim from `nodes.jsonl`'s `id` field). | | `/enrich --touched` | Per-PR incremental. Reads `lineage/{repo}/manifest.yaml`'s `enrichment.last_enriched_commit`; computes `git diff` against repo HEAD; enriches every node whose source file appears in the diff (1-hop fanout NOT YET — slice 5 is single-file granularity). |