go-skills-knowledge-pack
Skill0 starsChanged 3 months agoApache-2.0View on GitHub
Use this skill when designing, reviewing, or updating Go systems for projects: high-load crawlers, API backends, browser automation, AI agents, CRM/data/search platforms, messaging/workflows, TUI/CLI tools, and security-sensitive infrastructure. Prefer latest stable versions, pinned modules, official documentation, and strict CI/security checks.
In this repository78 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .agents/skills/go-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .agents/skills/rust-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .agents/skills/typescript-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/go-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/rust-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/typescript-skills-knowledge-pack/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/00-general.mdcCursor rule
- .cursor/rules/revharness-critical.mdcCursor rule
- .cursor/rules/revharness-detailed.mdcCursor rule
- auto-orchestratorSkill
- baseline-protectionSkill
- client-distribution-readinessSkill
- cloudflare-deploy-guardSkill
- codex-app-server-guardSkill
- codex-app-server-product-integrationSkill
- codex-callerSkill
- context-compactorSkill
- cursor-callerSkill
- delegated-development-disciplineSkill
- deploy-command-disciplineSkill
- design-principlesSkill
- development-constitutionSkill
- development-junk-cleanupSkill
- harness-official-docs-updateSkill
- naming-normalization-guardSkill
- orchestrator-bootstrapSkill
- payload-cms-deploy-guardSkill
- production-function-implementerSkill
- research-handoffSkill
- rev-harness-lifecycleSkill
- review-workflowSkill
- root-cause-first-disciplineSkill
- rust-skills-knowledge-packSkill
- scope-guardSkill
- self-growth-proposal-triageSkill
- shared-client-wrapper-safetySkill
- slice-designerSkill
- staff-code-reviewerSkill
- structured-mentorSkill
- supabase-deploy-guardSkill
- system-plannerSkill
- typescript-skills-knowledge-packSkill
- auto-orchestratorSkill
- baseline-protectionSkill
- client-distribution-readinessSkill
- cloudflare-deploy-guardSkill
- codex-app-server-guardSkill
- codex-app-server-product-integrationSkill
- codex-callerSkill
- context-compactorSkill
- cursor-callerSkill
- delegated-development-disciplineSkill
- deploy-command-disciplineSkill
- design-principlesSkill
- development-constitutionSkill
- development-junk-cleanupSkill
- go-skills-knowledge-packSkill
- harness-official-docs-updateSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
17 rules. About 0 are common, 13 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈13
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 54
Do not create http.Client per request.
- Very specific: line 55
Always set deadline/cancellation through context.Context.
- Distinctive: line 56
Use middleware for request id, auth, timeout, body size limit, logging, tracing.
- Distinctive: line 72
Treat browser automation as escalation, not default crawling.
- Distinctive: line 74
Never let queues become unbounded.
- Distinctive: line 88
Keep operational truth in PostgreSQL.
- Distinctive: line 89
Keep semantic retrieval in vector store or pgvector lane.
- Distinctive: line 90
Avoid huge eager loads. Stream rows and batch writes.
- Distinctive: line 107
Use idempotency keys for tool execution.
- Distinctive: line 121
Keep network/LLM calls outside render path.
- Distinctive: line 122
Use structured logs even for CLI tools.
- Distinctive: line 135
Do not invent cryptographic protocols casually.
- Distinctive: line 137
Keep secret material out of logs, traces, panics, and test fixtures.
- Very specific: line 153
Use pprof before “optimizing.”
- Distinctive: line 158
When Cloudflare, Supabase, Rust, TypeScript, and Go skills are loaded together:
- Very specific: line 166
Read references/multiskill-interop.md before making cross-language architecture decisions.
- Distinctive: line 170
Run in every serious update:
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
GoSkills
- Version:
v0.1.1-strict - Snapshot date: 2026-05-06 JST
- Generated from:
go-skills-master.md - Source registry:
go-skills-sources.md
この SKILL.md は Codex / Claude Code / ChatGPT Skills の軽量入口。判断の本体は go-skills-master.md、根拠URLは go-skills-sources.md、分野別詳細は references/ に置く。
Dependency version、採用レーン、例外、移行判断を出すときは、必ず go-skills-master.md を先に読む。SKILL.md だけでmodule registerを確定しない。
Mission
GoSkills は、のプロジェクト群に Go を投入するための実戦Skillである。RustSkills の低レイヤー性能思想と TypeScriptSkills のAI/フルスタック運用思想を、Goの標準ライブラリ・goroutine・単一バイナリ運用に写像する。
Default stance
- 最新安定版を使う。
go.mod/go.sum/ tool version を固定する。- v0.x module は「最新タグ」でも stable 扱いしない。Adopt/Watch/R&Dで扱う。
- 公式docs、pkg.go.dev、GitHub Releases、Go vulnerability database を優先する。
- 100点判定は実repoで
go test,govulncheck,gosec,golangci-lint,pprof, benchmark を通してから行う。
Primary toolchain
- Go: 1.26.6
- Production baseline: latest Go patch release
- CI: same Go patch + previous supported Go major if compatibility is required
- Security:
govulncheck,gosec,go mod verify - Lint:
go vet,staticcheck,golangci-lint
Architecture lanes
1. API / backend
Use:
net/httpas default HTTP coregithub.com/go-chi/chi/v5for routingconnectrpc.com/connectfor browser-compatible RPCgoogle.golang.org/grpcfor strict gRPC interopgithub.com/danielgtaylor/huma/v2for REST + OpenAPI 3.1
Rules:
- Do not create
http.Clientper request. - Always set deadline/cancellation through
context.Context. - Use middleware for request id, auth, timeout, body size limit, logging, tracing.
2. High-load crawler / form sender
Use:
net/http.Transporttuned per target classgolang.org/x/sync/errgroupwithSetLimitgolang.org/x/time/ratefor token bucket limitsgithub.com/cenkalti/backoff/v5for retry policygithub.com/gocolly/colly/v2for crawler abstractiongithub.com/chromedp/chromedp/github.com/go-rod/rodfor browser/CDP escalationgithub.com/valyala/fasthttponly for measured hot paths
Rules:
- Treat browser automation as escalation, not default crawling.
- Separate global concurrency, host-level concurrency, and account/campaign rate limits.
- Never let queues become unbounded.
3. Data / CRM / search
Use:
github.com/jackc/pgx/v5for PostgreSQLgithub.com/sqlc-dev/sqlcfor type-safe SQL codegenentgo.io/entfor schema graph/codegen where usefulgithub.com/qdrant/go-clientfor vector DB- pgvector/SQL path for PostgreSQL-native vector workflows when appropriate
Rules:
- Keep operational truth in PostgreSQL.
- Keep semantic retrieval in vector store or pgvector lane.
- Avoid huge eager loads. Stream rows and batch writes.
4. AI agents / MCP / workflows
Use:
github.com/openai/openai-go/v3for official OpenAI API accessgithub.com/modelcontextprotocol/go-sdkv1.6.0 as official MCP SDK adopt lanegithub.com/tmc/langchaingoas R&Dgithub.com/nats-io/nats.gofor event bus / JetStreamgo.temporal.io/sdkfor durable workflowsgithub.com/hibiken/asynqfor Redis task queue watch lane
Rules:
- Make agent tool calls typed and auditable.
- Persist durable state outside the LLM loop.
- Use idempotency keys for tool execution.
5. CLI / TUI
Use:
github.com/spf13/cobrafor CLIcharm.land/bubbletea/v2for TUI runtimecharm.land/lipgloss/v2for stylingcharm.land/bubbles/v2for components
Rules:
- Separate app state, update loop, renderer, and side effects.
- Keep network/LLM calls outside render path.
- Use structured logs even for CLI tools.
6. Security / E2EE
Use:
crypto/tlsand standard crypto firstgolang.org/x/cryptofor supplemental cryptogithub.com/cloudflare/circlfor PQ/ECC R&Dgithub.com/flynn/noisefor Noise Protocol R&D
Rules:
- Do not invent cryptographic protocols casually.
- Review nonce/counter/AAD/key rotation explicitly.
- Keep secret material out of logs, traces, panics, and test fixtures.
7. Observability / performance
Use:
log/slogdefaultgo.uber.org/zaporgithub.com/rs/zerologfor performance logginggo.opentelemetry.io/otelgithub.com/prometheus/client_golangruntime/pprof,runtime/metrics,testing.Bgo.uber.org/automaxprocsin containers
Rules:
- Measure p50/p95/p99 and allocations/op.
- Use
pprofbefore “optimizing.” - Watch GC pauses and goroutine counts under load.
8. Multi-skill platform integration
When Cloudflare, Supabase, Rust, TypeScript, and Go skills are loaded together:
- Cloudflare runtime rules override Go runtime assumptions for Workers.
- Supabase official/client rules override Go community-client assumptions.
- Rust owns low-level E2EE/performance kernels unless explicitly reassigned.
- TypeScript owns browser, frontend, and general Cloudflare Worker glue.
- Go owns backend APIs, CLIs, durable workflows, service workers, and orchestration.
Read references/multiskill-interop.md before making cross-language architecture decisions.
Required update checklist
Run in every serious update:
go version
go env GOVERSION
go list -m -u -json all
go mod tidy
go mod verify
go test ./...
go test -race ./...
go vet ./...
govulncheck ./...
gosec ./...
golangci-lint run
staticcheck ./...
go test -bench=. -benchmem ./...References
Detailed rules live in these reference files. They are lane-level expansions generated from the master, not independent policy:
references/runtime-toolchain.mdreferences/api-backend-rpc.mdreferences/highload-crawler-browser.mdreferences/data-persistence-search.mdreferences/ai-agents-mcp.mdreferences/tui-cli-audio.mdreferences/crypto-security-e2ee.mdreferences/messaging-workflows.mdreferences/performance-memory.mdreferences/observability-governance.mdreferences/testing-release.mdreferences/multiskill-interop.md
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.

