Skip to content

Commit b60700a

Browse files
committed
docs: clarify cluster reporting format in skill
1 parent 2fa7939 commit b60700a

2 files changed

Lines changed: 13 additions & 0 deletions

File tree

skills/ghcrawl/SKILL.md

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -145,11 +145,15 @@ ghcrawl neighbors owner/repo --number 42 --limit 10
145145
- Report the repo name and whether you refreshed data in this run.
146146
- When listing clusters, include:
147147
- cluster id
148+
- representative number and kind
148149
- display title
149150
- total size
150151
- PR count
151152
- issue count
152153
- latest updated time
154+
- When naming a cluster in prose, use this shape:
155+
- `Cluster <clusterId> (#<representativeNumber> representative <issue|pr>)`
156+
- example: `Cluster 23945 (#42035 representative issue)`
153157
- When drilling into a cluster, include clickable GitHub links for each issue/PR if you mention them.
154158
- Prefer concise summaries over dumping raw JSON.
155159
- If freshness is stale, say that explicitly:

skills/ghcrawl/references/protocol.md

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -72,6 +72,15 @@ Each cluster includes:
7272
- `representativeNumber`
7373
- `representativeKind`
7474

75+
When reporting a cluster to the user, do not mention only the cluster id. Use:
76+
77+
- `Cluster <clusterId> (#<representativeNumber> representative <issue|pr>)`
78+
79+
Examples:
80+
81+
- `Cluster 23945 (#42035 representative issue)`
82+
- `Cluster 104 (#38112 representative pr)`
83+
7584
This is the normal read-only exploration command for existing local data.
7685

7786
### `ghcrawl cluster-detail owner/repo --id <cluster-id>`

0 commit comments

Comments
 (0)