复制安装命令
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
Essays and writing behind this toolkit live at vexjoy.com.
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
来源文件:README.md
Essays and writing behind this toolkit live at vexjoy.com.
VexJoy Agent connects plain-English requests to specialist agents, skills, and workflows. /do selects the knowledge and tools needed for your task. Hooks enforce specific checks, and scripts handle repeatable work.
The aim is to give capable models useful domain knowledge without making you learn the toolkit's catalog.
43 domain agents, 59 workflow skills, 78 hooks, 153 scripts. Agents carry knowledge, skills enforce methodology, hooks block incomplete work, scripts handle determinism.
Works across Claude Code (/do), Codex ($do), Factory (/do), Reasonix (/do).
$ claude
> /do debug this Go test
Routing: go-engineer + systematic-debugging
Phase 1/4: Reproduce: running test, capturing failure...
Phase 2/4: Hypothesize: 3 candidates from stack trace...
Phase 3/4: Verify: isolated root cause in connection pool timeout
Phase 4/4: Fix: patch applied, test passing, PR opened
✓ Delivered: PR #847, fix connection pool timeout in health check
The router pairs a Go agent with a debugging skill, then follows the task through verification and delivery.
ROUTE PLAN EXECUTE VERIFY DELIVER RECORD
┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐
│ /do │───▶│ Task │───▶│Agent │───▶│Tests │───▶│ PR │───▶│Route │
│Router│ │ Plan │ │+Skill│ │Gates │ │Branch│ │Result│
└──────┘ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘
/d routes requests through TypeSafe's Jev classifier. One API call picks the agent, skill, and pipeline — no manifest read into context. Requires Jev; use /do if TypeSafe is not configured.
Setup: install the typesafe MCP plugin and set TYPESAFE_API_KEY in your environment.
> /d fix the flaky test in the payments module
ROUTING (/d): testing-automation-engineer + testing-preferred-patterns
Source: jev (confidence: medium)
Invoking...
Checks require evidence rather than confidence.
| Agent Says | What Happens |
|---|---|
| "Code looks correct, skip tests" | Exit gate requires test output. Blocked. |
| "Trivial change, no verification" | Hook blocks completion without evidence. |
| "Similar to before" | Skill demands case-specific proof. |
| "User is in a hurry" | Protocol overrides time pressure. |
| "I'm confident" | Gate demands exit code, not assertion. |
Hooks run at configured events. Skills state what to verify; blocking78 hooks enforce the checks they cover. Coverage depends on the runtime and tool path.
The content engine researches, drafts in a calibrated voice, checks 397 writing patterns, and adapts finished pieces for each platform. /html produces a self-contained report, slide deck, prototype, chart, or diagram. It needs no coding or setup beyond installation.
Toolkit changes use direct review and relevant checks. Model comparisons can settle specific uncertainties; they are not required for every edit. PHILOSOPHY.md explains the validation policy. what-didnt-work.md records failed experiments, routing reversals, unvalidated A/B citations, disabled lint rules, and program refutations.
The automated nightly evolution loop (/evolve, writes to evolution-reports/) ran regularly through mid-May 2026. It is currently dormant; recent evidence has come from manual PRs instead.
git clone https://github.com/notque/vexjoy-agent.git ~/vexjoy-agent
cd ~/vexjoy-agent
./install.sh
Installs into ~/.claude/ and mirrors into ~/.codex/, ~/.factory/, and ~/.reasonix/ when the runtime command is on PATH or its home directory exists. Choose symlinks for live updates through git pull, or copies for a stable snapshot.
Want only part of the toolkit? Run ./install.sh --configure to pick which skills, agents, and78 hooks install, or copy .local.example/profile.yaml to .local/profile.yaml and edit. No profile file = full install, unchanged behavior. Credit: @thomasvan. Details: .local.example/README.md.
| CLI | Entry Point |
|---|---|
| Claude Code | /do |
| Codex | $do |
| Factory | /do |
| Reasonix | /do |
Jev Auto-Compact plugin (optional, requires TYPESAFE_API_KEY):
claude plugin marketplace add ./plugins/jev-auto-compact
claude plugin install jev-auto-compact@jev-auto-compact -y
Replaces LLM-generated compaction summaries with Jev-judged verbatim pruning. Once context reaches 60%, Jev evaluates each old tool call (keep, truncate result, or drop) and returns the pruned transcript with zero rewriting, in about a second instead of one to three minutes. The threshold matters: every compaction is a cold KV-cache rewrite of the prefix, so compacting every turn multiplies cost. Evidence lives in learning.db (python153 scripts/jev-compact-evidence.py).
Proof it works: python153 scripts/jev-compact-evidence.py prints every compaction from two sources side by side — the plugin's claim and the engine's own compact_boundary record (tokens before/after, duration). A Jev compaction shows as a sub-second engine record next to a matching plugin claim; a built-in LLM compaction shows as a 30–150s record. Rows live in ~/.claude/learning/learning.db (compaction_events, session_usage).
Full setup: docs/start-here.md
Mirrors agents, skills, and supported78 hooks into ~/.codex/. The original six-hook allowlist was correct for Codex v0.114, when tool hooks only intercepted Bash. Current support requires Codex v0.144.1+ and classifies the 62 Claude hook registrations as 26 native, 27 adapter-backed, and 9 unsupported (53 supported). These are registration counts, not unique hook files. The installer also preserves explicit per-subagent model routing for GPT-5.6 Sol by setting the MultiAgent V2 compatibility keys documented in openai/codex#31814.
Codex now exposes apply_patch to tool78 hooks. VexJoy's adapter converts each patch operation into the Write/Edit payload expected by existing guards, but it cannot intercept writes performed through unified_exec, unmatched MCP tools, WebSearch, or other unsupported tool paths. PreCompact and Stop adapters also receive less telemetry than Claude Code: Codex does not provide Claude's conversation_history or session_data. This is expanded compatibility, not full Claude parity.
After install or any hook-definition change, run /hooks in Codex and review the new definitions before trusting them. Codex hash-trusts hook commands and skips changed, unreviewed definitions.
Gemini CLI support removed (deprecated upstream, transitioned to Antigravity CLI); Antigravity support pending CLI maturity. Per Google's transition announcement, Gemini CLI stops serving requests on 2026-06-18 for Google AI Pro / Ultra and free Gemini Code Assist for individuals. Gemini API integrations (image-gen backends, sprite pipeline, GEMINI_API_KEY) are unaffected and stay in the toolkit.
If a prior install mirrored into ~/.gemini/, remove the stale mirrors with:
rm -rf ~/.gemini/skills ~/.gemini/agents ~/.gemini/hooks ~/.gemini/scripts ~/.gemini/antigravity/plugins/vexjoy-agent
Mirrors agents (as "droids"), skills, and all78 hooks into ~/.factory/. Hook config merges into ~/.factory/settings.json with paths rewritten.
Mirrors skills, 153 scripts, and the allowlisted 78 hooks (scripts/reasonix-hooks-allowlist.txt) into ~/.reasonix/ (no agent or custom-command surface, so neither is installed; the /do router rides in as a skill). Reasonix fires only 4 events (PreToolUse, PostToolUse, UserPromptSubmit, Stop), so only hooks for those events are allowlisted. Hook config is written to the hooks key of ~/.reasonix/settings.json in Reasonix's native flat shape (one entry per hook, match regex over the tool name); the generator builds absolute python3 commands, so no path rewrite is applied. MCP/model/permissions in ~/.reasonix/config.json are user-owned and left untouched.
The toolkit supplies its own routing, domain knowledge, methodology, and enforcement. The default system prompt duplicates most of that.
claude --system-prompt "."
Strips built-in tool-use instructions. The toolkit's agents, skills,78 hooks, and CLAUDE.md provide equivalent coverage.
| Layer | Count | Does |
|---|---|---|
| Agents | 43 | Domain knowledge: idiom tables, failure mode catalogs, error-to-fix mappings |
| Skills | 59 | Phased methodology with gates. Can't skip steps. Each phase has exit criteria requiring evidence. |
| Hooks | 78 | Fire on lifecycle events. Block incomplete work. Zero LLM cost. |
| Scripts | 153 | Determinism: test runners, linters, validators. No LLM judgment. |
Full skill catalog: docs/skills.md.
┌─────────────────────────────────────────────────┐
│ SKILL.md │
│ ┌─ Frontmatter ─────────────────────────────┐ │
│ │ triggers, pairs_with, success-criteria │ │
│ └────────────────────────────────────────────┘ │
│ Reference Loading Table (conditional imports) │
│ Phased Instructions (numbered, with gates) │
│ Verification (evidence requirements) │
└─────────────────────────────────────────────────┘
A game built entirely by Claude Code using these agents, skills, and pipelines:
I just want to use it Install, learn /do, done.
I do knowledge work Writing, research, data analysis, moderation, HTML artifacts. No code.
I'm a developer Architecture, extension points, adding agents and skills.
I'm an AI power user Routing tables, pipelines,78 hooks, telemetry DB.
I'm an AI agent Machine-dense inventory. Tables, paths, schemas.
I'm on LinkedIn 🚀 Thought leadership. Agree? 👇
Full design philosophy: PHILOSOPHY.md
One report-only script surfaces upkeep work; it prints a digest and never edits, deletes, or blocks.
python153 scripts/stale-skill-scan.py --top 20 ranks stale skills and agents as pruning candidates. Run it quarterly; see docs/deprecation-template.md.Scheduled work follows the same boundary as everything else: judgment uses agents; repeatable plumbing uses153 scripts.
| Need | Use |
|---|---|
| Run a deterministic command on a schedule | scripts/agent-scheduler.py with runner: "command" |
| Run an agent judgment on a schedule, webhook, or file change | scripts/agent-scheduler.py with the default runner: "claude" |
| Install or remove a user crontab entry safely | scripts/crontab-manager.py |
| Audit shell cron reliability | cron-automation |
| Keep one interactive objective moving until criteria verify | objective-loop |
See CONTRIBUTING.md.
MIT. See LICENSE.
name: programming
description: "Language-specific patterns and tooling: Go, Kotlin, PHP, Swift, TypeScript."
user-invocable: false
allowed-tools:
- Read
- Write
- Bash
- Grep
- Glob
- Edit
- Task
- Skill
routing:
force_route: true
not_for: "Python (agents handle directly), general debugging (use debugging). When a dedicated language skill exists (php, swift, kotlin), prefer it over this umbrella"
triggers:
# Go
- .go
- go test
- "*_test.go"
- goroutine
- sync.Mutex
- context.Context
- Go code review
- Go lint
- sapcc
- make check
# Kotlin
- "kotlin coroutines"
- "kotlin Flow"
- "suspend function"
- "kotlin testing"
- kotest
# PHP
- "php quality"
- "php code review"
- "PSR standards"
- phpstan
- phpunit
# Swift
- "swift concurrency"
- "swift async await"
- "Swift Actor"
- XCTest
# TypeScript
- "TypeScript check"
- "tsc noEmit"
- "tsc errors"
category: language
pairs_with:
- golang-general-engineer
- programming-general-engineer
- typescript-frontend-engineerLanguage-specific patterns and tooling for Go, Kotlin, PHP, Swift, and TypeScript. Identify the language, load matching deep references, follow the language section.
| Language | Agent | Section |
|---|---|---|
| Go | golang-general-engineer | Go |
| Kotlin | kotlin-general-engineer | Kotlin |
| PHP | php-general-engineer | PHP |
| Swift | swift-general-engineer | Swift |
| TypeScript | typescript-frontend-engineer | TypeScript |
If the request spans multiple languages, apply each relevant section.
Prefer local consistency only where the Google guide allows judgment. Do not restyle existing code without cause.
For testing: table-driven tests with subtests, test helpers, mock injection. Run go test ./.... Load references/go/testing.md.
For concurrency: goroutine lifecycle, sync primitives, context propagation, race detection. Run go test -race. Load references/go/concurrency.md.
For error handling: wrapping with %w, sentinel errors, errors.Is/errors.As, custom error types. Load references/go/error-handling.md.
For code review: review checklist, common comments, interface compliance. Load references/go/code-review.md.
For SAPCC repos: org conventions -- go-bits patterns, auth ordering, JSON strictness, endpoint identification. Load references/go/sapcc-conventions.md.
For quality gate: run make check or equivalent. Load references/go/quality-gate.md.
Beyond linting -- patterns from real code reviews that automated tools miss:
commit() should remove only the specific batch returned, not all items.defer f.Close() after the error check, not before.i := i reassignment and closure parameter passing -- each iteration has its own variable since 1.22..Add(0) for all known label combinations.testWithEachBackingStore() pattern to share test bodies across interface implementations.| Task | Script |
|---|---|
| Generate table test | scripts/gen-table-test.sh |
| Benchmark compare | scripts/bench-compare.sh |
| Check errors | scripts/check-errors.sh |
| Interface compliance | scripts/check-interface-compliance.sh |
| SAPCC checks | scripts/check-sapcc-*.sh |
| Quality checker | scripts/quality_checker.py, scripts/validate.py |
| Google style sync | scripts/sync-google-style-guide.py |
CoroutineDispatcher as a parameter so callers and tests control threading.Exception.| Type | Temperature | Replay | Use Case |
|---|---|---|---|
Flow | Cold | None | Async sequences, API pagination, transform chains |
StateFlow | Hot | Latest (1) | UI state with current value; conflates duplicates |
SharedFlow | Hot | Configurable | One-shot events: navigation, toasts, errors |
Channel | Hot | None | Producer-consumer, fan-in/fan-out inter-coroutine communication |
Flow: cold -- each collector gets its own execution. Chain with filter, map, flatMapConcat. Terminate with collect.
StateFlow: requires initial value, always replays latest to new collectors. Use for UI state.
SharedFlow: configurable replay, emits all values (no conflation). Use for event streams.
Channel: hot communication primitive. Use produce builder for producer-consumer. Fan-out: multiple coroutines consuming one channel. Fan-in: multiple producers writing to one channel.
declare(strict_types=1).Laravel: Eloquent scopes for reusable query constraints (scopeActive). Collection methods over raw loops (collect()->filter()->map()->sort()). Service Container for interface-to-implementation binding. Contextual binding with when()->needs()->give().
Symfony: DI with #[Autowire] attributes for constructor injection. #[AsEventListener] for event dispatch. Rely on autowiring over XML config.
TestCase with test prefix, @dataProvider for table-driven, createStub() for returns, createMock() for interactions, parent::setUp() first in every setUp../vendor/bin/phpunit. For coverage: XDEBUG_MODE=coverage ./vendor/bin/phpunit --coverage-text --coverage-min=80.Concurrency:
TaskGroup over loose Task { }.Sendable. Enable -strict-concurrency=complete.@MainActor for UI layer only. Use nonisolated for methods that only read let properties.Task needs a cancellation path.Testing:
testFetchUser_withExpiredToken_throwsAuthError.@Test, #expect) for Swift 5.9+. Fall back to XCTest for older targets or UI tests.arguments:. Protocol-based mock injection.swift test --enable-code-coverage and swift build. Confirm strict concurrency checking passes.Read-only type validation. Do not fix code, change configuration, or install dependencies for a check-only request.
tsconfig.json (check src/, app/, packages/). In a monorepo, use the affected package or the repo's aggregate check. If no config exists, report and stop.npx tsc --noEmit 2>&1 (or --project path/to/tsconfig.json). Capture exit code. Zero = pass; nonzero = type errors or tool failure (report which).file:line:column, TS####, and the message.npm install typescript --save-dev.npm exec tsc -- --noEmit.--skipLibCheck: use only when configured or requested.--strict: enable only when configured or requested.--incremental: avoid for read-only checks (writes build-info cache).For the full JS/TS check sequence: lint first, type-check next, then run tests.
Load only the references matching the task. Multiple may apply (e.g., a concurrency PR review loads both code-review and concurrency references).
| Signal | Reference |
|---|---|
| Tests, table-driven, helpers, mocks | references/go/testing.md + sub-refs in testing/ |
| Goroutines, channels, sync, race conditions | references/go/concurrency.md + concurrency/concurrency-patterns.md |
| Error wrapping, sentinels, errors.Is/As | references/go/error-handling.md + error-handling/patterns.md |
| Code smells, bad patterns | references/go/preferred-patterns.md + preferred-patterns/code-examples.md |
| Code review process | references/go/code-review.md + code-review/common-review-comments.md |
| SAPCC repos, go-bits | references/go/sapcc-conventions.md + sub-refs in sapcc-conventions/ |
| make check, linting | references/go/quality-gate.md + sub-refs in quality-gate/ |
| Google style decisions | references/go/google-style-guide/ (pinned upstream) |
| Signal | Reference |
|---|---|
| Scopes, cancellation, dispatchers, exception handling | references/kotlin/concurrency-patterns.md |
| JUnit 5, Kotest, MockK, runTest | references/kotlin/kotlin-testing.md |
| Signal | Reference |
|---|---|
| Union/intersection types, enums, readonly, match | references/php/modern-php-features.md |
| PHP-CS-Fixer, PHPStan, Psalm, Rector | references/php/quality-tools.md |
| PHPUnit, data providers, mocks, stubs | references/php/testing-patterns.md |
| Signal | Reference |
|---|---|
| TaskGroup, AsyncSequence, AsyncStream, cancellation | references/swift/task-patterns.md |
| XCTest, Swift Testing, async tests, UI tests | references/swift/swift-testing.md |
评论 (0)
暂无评论,成为第一个评论者吧!