Best for
- Use when asked to triage the backlog, clean up issues, find duplicates, or report on issue health.
dfrostar/neuralmind/skills/github-issue-audit/SKILL.md
Audit a GitHub repository's open issues for hygiene — stale issues, likely duplicates, missing/incorrect labels, and issues that are actually already fixed on main. Use when asked to triage the backlog, clean up issues, find duplicates, or report on issue health. Produces a ranked, actionable report; it never closes or edits issues on its own.
Decision brief
Produce an honest, actionable audit of a repo's open issues. You read and recommend; you do not close, relabel, or comment on issues unless the user explicitly asks for that action after seeing the report.
Compatibility matrix
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/dfrostar/neuralmind --skill "skills/github-issue-audit"Inspect the Agent Skill "github-issue-audit" from https://github.com/dfrostar/neuralmind/blob/943df635efb24992b4c53470e856a4438eb1c6cb/skills/github-issue-audit/SKILL.md at commit 943df635efb24992b4c53470e856a4438eb1c6cb. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
Be frugal with API calls — batch, use minimaloutput when the full body isn't needed, and only deep-read issues that a cheap signal already flagged.
Four checks, run over every open issue, reported most-actionable first:
A markdown table, then a short "recommended actions" list. Rank by actionability (fixed duplicate mislabeled stale), not by issue number.
If the user says "go ahead and apply these" (or similar), then and only then: mutate in small, reversible steps — relabel before closing, close with a comment that references the duplicate/fix, and never bulk-close on medium confidence without a second confirmation. Report what…
Permission review
No configured static risk pattern was detected
This is not proof of safety. Runtime behavior, indirect dependencies, and hidden external systems are outside the static scan.
Evidence record
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 75/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 22 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Produce an honest, actionable audit of a repo's open issues. You read and recommend; you do not close, relabel, or comment on issues unless the user explicitly asks for that action after seeing the report.
Four checks, run over every open issue, reported most-actionable first:
main. Cross-check
with list_commits / search_code before claiming this; a fixed-looking
issue that's actually still open is the most embarrassing miss.1. get_me confirm permissions/context
2. list_issues (state=open) page in batches of 5-10, minimal_output
3. For each issue:
issue_read full body + current labels + timeline
classify into the 4 buckets above
4. For "already fixed" candidates only:
search_code / list_commits verify the fix actually landed
5. Emit the report (below). Stop. Do not mutate anything.
Be frugal with API calls — batch, use minimal_output when the full body
isn't needed, and only deep-read issues that a cheap signal already flagged.
A markdown table, then a short "recommended actions" list. Rank by actionability (fixed > duplicate > mislabeled > stale), not by issue number.
| # | Title | Finding | Confidence | Suggested action |
|---|-------|---------|------------|------------------|
| 142 | "crash on empty repo" | already fixed in a1b2c3d | high | close w/ ref |
| 130 | "add --json flag" | dup of #98 | medium | link + close one |
Every "close" or "relabel" suggestion is a proposal. Confidence is honest:
high only when you verified (a commit ref, an exact-duplicate body);
medium/low otherwise, and say what would raise it.
If the user says "go ahead and apply these" (or similar), then and only then:
mutate in small, reversible steps — relabel before closing, close with a
comment that references the duplicate/fix, and never bulk-close on medium
confidence without a second confirmation. Report what you changed.
get_label / the labeler config first.Alternatives
coreyhaines31/marketingskills
When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program
coreyhaines31/marketingskills
When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' 'involuntary churn,' 'people keep canceling,' 'churn rate is too high,' 'how do I keep users,' or 'customers are leaving.' Use this whenever someone is losing subscribers o
event4u-app/agent-config
Use when the user says "review the design", "check the UI", or wants a comprehensive UI/UX review. Uses a 7-phase methodology covering interaction, responsiveness, accessibility, and more.
K-Dense-AI/scientific-agent-skills
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.