gh-issues-by-severity
Skill4 starsChanged 8 days agoMITView on GitHub
List open GitHub issues for the current repo grouped by severity, highest first, using the repo's own severity:* labels. Use when the user asks to see open issues by severity or wants a standup-style issue scan.
In this repository15 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- cavemanSkill
- grid-helpSkill
- grill-meSkill
- growth-psychSkill
- handoffSkill
- mine-learningsSkill
- openspec-helpSkill
- rubber-duckSkill
- setup-repo-skillsSkill
- skill-scoutSkill
- spec-scoutSkill
- standupSkill
- tightenSkill
- un-claudishSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
If this is empty, the repo has no severity scheme — every issue falls into a single unlabeled bucket.
- Very specific: line 81
If gh isn't authenticated or the directory isn't a GitHub repo, surface the gh error directly rather than guessing at a fix.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
gh-issues-by-severity
List open issues, grouped by severity (most severe first), sorted by issue
number ascending within each group. Generic across repos: discovers whatever
severity:* labels the repo actually has rather than assuming a fixed set.
Process
- Discover the severity label vocabulary for this repo:
If this is empty, the repo has no severity scheme — every issue falls into a singlegh label list --json name --jq '[.[].name | select(startswith("severity:"))]'unlabeledbucket. Say so, then continue (still produce the list, sorted by issue number). Rank the discovered labels. Strip the
severity:prefix and match each value (case-insensitive) against these clusters, highest severity first:critical,blocker,p0,sev1high,major,p1,sev2medium,moderate,p2,sev3low,minor,trivial,p3,p4,sev4
Any discovered value that matches none of these clusters is "unrecognized" — rank it below all recognized clusters, ordered alphabetically among other unrecognized values.
unlabeled(noseverity:*label on the issue) always ranks lowest of all.Build a rank map, e.g. for a repo with
critical/high/medium/low:{"critical": 4, "high": 3, "medium": 2, "low": 1, "unlabeled": 0}- Fetch open issues and sort server-side via jq — don't hand-sort the
list yourself:
(If the user named a different repo, addgh issue list --state open --limit 1000 --json number,title,labels \ | jq --argjson rank '<rank map from step 2>' ' map({ number, title, severity: ((.labels[] | select(.name | startswith("severity:")) | .name | sub("^severity:";"")) // "unlabeled"), labels: [.labels[].name | select(startswith("severity:") | not)] }) | map(. + {rank: ($rank[.severity] // -1)}) | sort_by([-.rank, .number]) '--repo <owner>/<name>.) Render the sorted JSON as grouped markdown, one section per severity level present (in descending rank order), with a count in the header:
## critical (2) #42 Auth bypass on login [bug, needs-triage] #58 Data loss on export [bug] ## high (1) #61 Slow query on dashboard [enhancement, needs-info] ## low (3) #12 Typo in README [] #19 Flaky test in CI [needs-triage] #30 Missing alt text []Show every label except the
severity:*one — no further filtering. Omit the trailing[]bracket if an issue has no other labels, or leave it empty; either is fine, just be consistent within one run.
Notes
- 1000-issue limit is for completeness on typical repos; if a repo somehow exceeds it, mention that the list may be truncated.
- If
ghisn't authenticated or the directory isn't a GitHub repo, surface thegherror directly rather than guessing at a fix.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

