Github Scanning

github-scanning Skill

GitHub data collection patterns for workflow agents. Covers search query construction by intent, date range handling, repository scope narrowing, preferences.md integration, cross-repo intelligence, parallel stream collection model, and auto-recovery for empty results.

Agents That Use This Skill

Agent	Why
github-hub	Routing and repo-wide discovery
daily-briefing	Multi-stream parallel collection
issue-tracker	Issue search queries
pr-review	Pull request search queries
analytics	Repository-level data collection
insiders-a11y-tracker	Accessibility commit and release tracking

Search Query Patterns by Intent

Issues

Intent	Query Pattern
Assigned to you	is:open assignee:USERNAME
@mentioned	is:open mentions:USERNAME
Authored by you	is:open author:USERNAME
Specific repo	repo:owner/repo is:open
Org-wide	org:ORGNAME is:open
Closed (for recap)	is:closed author:USERNAME

Pull Requests

Intent	Query Pattern
Awaiting your review	review-requested:USERNAME state:open
Your open PRs	author:USERNAME state:open
You reviewed, check updates	reviewed-by:USERNAME state:open
Org-wide	org:ORGNAME state:open

Date Range Handling

User Says	GitHub Qualifier
"last week"	created:>YYYY-MM-DD (7 days ago)
"this month"	created:>YYYY-MM-01
"today"	closed:YYYY-MM-DD
"between X and Y"	created:X..Y
Not specified	updated:>YYYY-MM-DD (30 days) - mention the assumption

Scope Narrowing

Scope	Qualifier
Single repo	repo:owner/name
All org repos	org:orgname
All user repos	user:username
Everything (default)	No qualifier

Preferences File Integration

Read .github/agents/preferences.md before searching. Key settings:

Preference Key	Effect
repos.discovery	Default scope: all / starred / owned / configured / workspace
repos.include	Always include these repos
repos.exclude	Always skip these repos
repos.overrides	Per-repo tracking toggles and label/path filters
search.default_window	Default time range when user doesn't specify
briefing.sections	Which sections to include in the daily briefing

Parallel Stream Collection

Run independent streams simultaneously. Don't serialize calls with no dependencies.

Daily Briefing - 3 Batches

Batch	Streams (run in parallel)
1	Issues + Pull Requests + Releases + Accessibility updates
2	Discussions + CI/CD health + Security alerts
3	Project board status + Recently closed/merged work

Batch 2 can overlap with Batch 3. Both depend on Batch 1 completing first.

Analytics - 2 Batches

Batch	Streams
1	Closed PRs + Opened issues + CI runs + Security alerts (all independent)
2	Health score + Velocity + Bottleneck detection (depend on Batch 1)

Announcement Template

 Running N searches in parallel...
 Batch 1 complete - X items found

 Running N additional searches...
 All complete - Y items collected

Cross-Repo Intelligence

When results arrive from multiple repos, surface these patterns:

Pattern	Action
Cross-repo references	Fetch referenced item and surface both together
Shared label patterns	Group P0 items from different repos
Merged PR not yet released	Note: "PR #N is merged but not in any release yet"
Issue with merged fix PR	Flag: "This may be resolved - PR #N that closes it was merged on {date}"

Auto-Recovery for Empty Results

Never return 0 results without trying at least one broader query first:

1. Remove date qualifier
2. Expand scope (add org: or remove repo: qualifier)
3. Remove label filters
4. Report 0 only after all three attempts return nothing

Always tell the user what was broadened: "No results in last 7 days - broadened to last 30 days and found 3 items."

Pagination Rules

• Batch results in groups of 10
• Always disclose total count: "Showing 10 of 47. Load more?"
• Never silently truncate

Deduplication

When the same item appears in multiple streams, show it once with all signals combined:
"Assigned, @mentioned" - priority = highest-scoring signal only, not additive.

Skill Location

.github/skills/github-scanning/SKILL.md