一键导入
github
REQUIRED --repo flag for all commands, helper functions for code review comments, suggestion block syntax rules
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
REQUIRED --repo flag for all commands, helper functions for code review comments, suggestion block syntax rules
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Cloudflare GraphQL Analytics for zone traffic, firewall events, Workers metrics, and schema exploration. Use when querying Cloudflare analytics data or exploring the GraphQL API.
PostgreSQL database analysis, performance tuning, and health monitoring. You MUST read this entire skill document before executing any PostgreSQL operations — it contains mandatory workflows, safety constraints, and two-phase execution rules that prevent common errors like hallucinated column names and unsafe queries.
SonarQube code quality and security analysis. Use when working with code quality metrics, security hotspots, quality gates, or issue tracking in SonarQube Cloud or Server.
Use when working with Aws Billing — analyze, break down, and report AWS costs and bills. Covers cost breakdown by service, account, or usage type; monthly/daily billing trends; cost anomaly detection; RI/SP utilization; cost forecasting; credit/discount analysis; and multi-account cost comparison. Uses anti-hallucination rules, mandatory currency/credit detection workflow, and reusable Cost Explorer functions.
Use when working with Aws Idle Resources — detect unused and idle AWS resources that incur cost without providing value. Covers detached EBS volumes, idle load balancers, unused Elastic IPs, stopped EC2 instances, idle NAT Gateways, old snapshots, and unused ENIs. Includes estimated monthly waste per resource and anti-hallucination rules for safe detection.
Use when working with Aws Pricing — aWS pricing helper for cost queries. ALWAYS use get_aws_cost script for pricing questions. Use when: - User asks about AWS resource costs or pricing - User wants to compare pricing across regions - User needs spot, on-demand, or reserved pricing info Triggers: aws pricing, aws cost, how much does, ec2 price, rds cost, s3 pricing.
| name | github |
| description | REQUIRED --repo flag for all commands, helper functions for code review comments, suggestion block syntax rules |
| connection_type | github |
| preload | false |
Execute GitHub CLI and git commands with proper authentication.
You must source the helpers before use:
source ./_skills/connections/github/github/scripts/github_helpers.sh
ALL commands require explicit --repo owner/repo flag.
✅ CORRECT: gh pr list --repo owner/repo
❌ WRONG: gh pr list # No repo context!
ALWAYS use github_create_pr_comment helper function for code review inline comments.
Why? Helper functions auto-handle:
Inline comment with suggestion block:
github_create_pr_comment OWNER REPO PR "file.py" LINE 'Comment text
```suggestion
fixed code
' "SIDE"
- `SIDE`: "RIGHT" for new code, "LEFT" for old code
- Use single quotes to preserve backticks literally - NO escaping needed
**Discussion comment:**
```bash
github_create_pr_discussion OWNER REPO PR 'Comment text'
✅ CORRECT - Single quotes preserve backticks:
github_create_pr_comment owner repo 123 "file.py" 45 '🔴 CRITICAL: Issue
```suggestion
if not input:
raise ValueError("Required")
' "RIGHT"
**❌ WRONG - Escaped backticks break rendering:**
```bash
github_create_pr_comment owner repo 123 "file.py" 45 "\`\`\`suggestion
code
\`\`\`" "RIGHT"
Key rules:
```suggestionSupported flags:
--repo owner/repo : REQUIRED - Repository target--state open|closed|merged|all : Filter by state--author username : Filter by author--assignee username : Filter by assignee--label name : Filter by label--limit N : Max number of PRs (default: 30)NOT SUPPORTED:
--created : Date filtering not available in gh CLIgh api with GraphQL for date filteringSupported flags:
--repo owner/repo : REQUIRED--state open|closed|all--author username--assignee username--label name--limit NNOT SUPPORTED:
--created-after : Not availablegh api endpointSupported flags:
--repo owner/repo : REQUIRED--json FIELDS : Output specific fields as JSON--web : Open in browserNOT SUPPORTED:
--diff : Not available (use gh pr diff instead)Supported flags:
--repo owner/repo : REQUIRED--patch : Show full patch formatNOT SUPPORTED:
--filter : File filtering not available (pipe to grep instead)Required parameters:
owner : Repository owner (org or user)repo : Repository namepr_number : PR number (not prefixed with #)file_path : File path in repositoryend_line : Line number for commentbody : Comment text (use single quotes for markdown)side : "RIGHT" (new code) or "LEFT" (old code), optional, default: RIGHTstart_line : Start line for multi-line, optionalNOT SUPPORTED:
commit_sha : Automatically fetched by helperposition : Automatically calculated by helperExample:
# ✅ Correct - single line comment
github_create_pr_comment owner repo 123 "src/main.py" 45 'Fix this issue' "RIGHT"
# ✅ Correct - multi-line with suggestion
github_create_pr_comment owner repo 123 "src/main.py" 45 '```suggestion
fixed code
```' "RIGHT" 43
# ❌ Wrong - escaped backticks
github_create_pr_comment owner repo 123 "src/main.py" 45 "\`\`\`suggestion..." "RIGHT"
Required parameters:
owner : Repository ownerrepo : Repository namepr_number : PR numberbody : Comment textBEFORE using gh CLI or GitHub helpers:
gh pr list --help to checkgh api with GraphQLCommon mistake pattern:
# User asks: "List PRs created after January 1"
# ❌ WRONG: Assume date filtering exists
gh pr list --repo owner/repo --created-after "2026-01-01" # Will fail
# ✅ CORRECT: Check SKILL.md first, use alternative
gh api graphql -f query='query { repository(owner:"owner", name:"repo") { pullRequests(first:100, orderBy:{field:CREATED_AT, direction:DESC}) { nodes { number title createdAt } } } }' | jq '.data.repository.pullRequests.nodes[] | select(.createdAt >= "2026-01-01")'
github_create_pr_commentPresent results as a structured report:
Github Report
═════════════
Resources discovered: [count]
Resource Status Key Metric Issues
──────────────────────────────────────────────
[name] [ok/warn] [value] [findings]
Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]
Target ≤50 lines of output. Use tables for multi-resource comparisons.
--help output.| Shortcut | Counter | Why |
|---|---|---|
| "I'll skip discovery and check known resources" | Always run Phase 1 discovery first | Resource names change, new resources appear — assumed names cause errors |
| "The user only asked for a quick check" | Follow the full discovery → analysis flow | Quick checks miss critical issues; structured analysis catches silent failures |
| "Default configuration is probably fine" | Audit configuration explicitly | Defaults often leave logging, security, and optimization features disabled |
| "Metrics aren't needed for this" | Always check relevant metrics when available | API/CLI responses show current state; metrics reveal trends and intermittent issues |
| "I don't have access to that" | Try the command and report the actual error | Assumed permission failures prevent useful investigation; actual errors are informative |