Skip to content
Back to search

vibecosystem

SkillWorks with Claude CodeGitHub

Profile-based Claude runtime with 138 agents, 296 skills, 74 hooks in the catalog; memory and orchestration are opt-in

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/vibeeval-vibecosystem/badge.svg)](/listings/vibeeval-vibecosystem)

HTML

<a href="/listings/vibeeval-vibecosystem"><img src="/api/listings/vibeeval-vibecosystem/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Use a source checkout to try these changes. When reviewing a PR, check out its head branch before installing; cloning uses `main` by default:

git clone https://github.com/vibeeval/vibecosystem.git
cd vibecosystem
./install.sh --profile core --dry-run
./install.sh --profile core

The installer uses the prebuilt hooks in `hooks/dist/`; a normal installation does not compile TypeScript or install the hook development dependencies. Node.js 18+ and Bash are required. Development checks use Node.js 22.

For the published npm version (currently 3.5.0):

npx vibecosystem init

The remote installer fetches `main` (this source tree). Review the script before running it:

curl -fsSL https://raw.githubusercontent.com/vibeeval/vibecosystem/main/install-remote.sh -o install-remote.sh
bash install-remote.sh --profile core

Restart Claude Code after changing its hook registrations. Inspect the installed state with:

vibeco effective-config
vibeco doctor
vibeco stats

`vibeco effective-config` shows which profile is active and how many hooks are registered and compiled.

The source installer creates `~/.local/bin/vibeco` unless an unmanaged launcher already exists there. Add `~/.local/bin` to `PATH` if your shell cannot find it. You can also run the source CLI with `node tools/vibeco/vibeco.mjs`.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

plugin.json · Manifest · 21 lines
  1. {
  2. "name": "vibecosystem",
  3. "description": "Profile-based Claude runtime with 138 agents, 296 skills, 74 hooks in the catalog; memory and orchestration are opt-in",
  4. "version": "3.5.1",
  5. "author": {
  6. "name": "vibeeval",
  7. "url": "https://github.com/vibeeval"
  8. },
  9. "repository": "https://github.com/vibeeval/vibecosystem",
  10. "license": "MIT",
  11. "keywords": [
  12. "claude-code",
  13. "agents",
  14. "skills",
  15. "hooks",
  16. "ai-team",
  17. "self-learning",
  18. "code-review",
  19. "tdd"
  20. ]
  21. }
README.md · README · 189 lines
  1. <div align="center">
  2. # vibecosystem
  3. **A focused runtime for Claude Code. A bounded worker for Codex.**
  4. [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
  5. **Catalog (in repo):** [![Agents](https://img.shields.io/badge/agents-138-blue.svg)](#catalog) [![Skills](https://img.shields.io/badge/skills-296-green.svg)](#catalog) [![Hooks](https://img.shields.io/badge/hooks-74-orange.svg)](#catalog) [![Rules](https://img.shields.io/badge/rules-20-red.svg)](#catalog)
  6. **Default core:** [![default core: 12 agents · 32 skills · 6 hooks](https://img.shields.io/badge/default%20core-12%20agents%20%C2%B7%2032%20skills%20%C2%B7%206%20hooks-green.svg)](#catalog)
  7. [![Validate](https://github.com/vibeeval/vibecosystem/actions/workflows/validate.yml/badge.svg)](https://github.com/vibeeval/vibecosystem/actions/workflows/validate.yml)
  8. [Türkçe](#turkce) · [Español](docs/README_ES.md) · [Français](docs/README_FR.md) · [Deutsch](docs/README_DE.md) · [Português](docs/README_PT.md) · [Italiano](docs/README_IT.md) · [Nederlands](docs/README_NL.md) · [中文](docs/README_ZH.md) · [日本語](docs/README_JA.md) · [한국어](docs/README_KO.md) · [العربية](docs/README_AR.md) · [हिन्दी](docs/README_HI.md) · [Русский](docs/README_RU.md)
  9. ![Core runtime by default; quality, context, memory and orchestration are explicit opt-ins](assets/runtime-demo.gif)
  10. </div>
  11. vibecosystem packages agents, skills and hooks for software work. The default `core` profile installs a selected set for investigation, implementation, review and verification. Memory, learning and orchestration have explicit profiles; `full` installs all 138 agents, 296 skills and 20 rules plus 53 of the 74 hooks.
  12. **3.5.1 is the current npm release.** See [CHANGELOG.md](CHANGELOG.md) and the [upgrade guide](UPGRADING.md).
  13. ## Quick start
  14. Use a source checkout to try these changes. When reviewing a PR, check out its head branch before installing; cloning uses `main` by default:
  15. ```bash
  16. git clone https://github.com/vibeeval/vibecosystem.git
  17. cd vibecosystem
  18. ./install.sh --profile core --dry-run
  19. ./install.sh --profile core
  20. ```
  21. The installer uses the prebuilt hooks in `hooks/dist/`; a normal installation does not compile TypeScript or install the hook development dependencies. Node.js 18+ and Bash are required. Development checks use Node.js 22.
  22. For the published npm version (currently 3.5.0):
  23. ```bash
  24. npx vibecosystem init
  25. ```
  26. The remote installer fetches `main` (this source tree). Review the script before running it:
  27. ```bash
  28. curl -fsSL https://raw.githubusercontent.com/vibeeval/vibecosystem/main/install-remote.sh -o install-remote.sh
  29. bash install-remote.sh --profile core
  30. ```
  31. Restart Claude Code after changing its hook registrations. Inspect the installed state with:
  32. ```bash
  33. vibeco effective-config
  34. vibeco doctor
  35. vibeco stats
  36. ```
  37. `vibeco effective-config` shows which profile is active and how many hooks are registered and compiled.
  38. The source installer creates `~/.local/bin/vibeco` unless an unmanaged launcher already exists there. Add `~/.local/bin` to `PATH` if your shell cannot find it. You can also run the source CLI with `node tools/vibeco/vibeco.mjs`.
  39. ## Profiles
  40. Profiles select installed components and registered hook commands. Catalog size does not describe how much prompt content is loaded into each request.
  41. | Profile | Purpose |
  42. | --- | --- |
  43. | `core` | Selected implementation, investigation and verification tools; default |
  44. | `dashboard` | Core plus opt-in local activity monitoring |
  45. | `quality` | Core plus focused edit, compiler and security checks |
  46. | `context` | Core plus targeted code context retrieval |
  47. | `memory` | Core plus persistent recall and learning |
  48. | `orchestration` | Core plus explicit multi-agent workflows |
  49. | `full` | All 138 agents, 296 skills and 20 rules; 53 of the 74 hooks |
  50. `minimal` and `smart` alias `core`; `all` aliases `full`. Legacy `frontend`, `backend`, `fullstack` and `devops` profiles retain domain allowlists with core hooks. Read the exact selection in [profiles/runtime-manifest.json](profiles/runtime-manifest.json), or use `vibeco effective-config` after installation.
  51. ```bash
  52. vibeco profile quality # core plus focused edit, compiler and security checks
  53. vibeco profile core # default selected set
  54. vibeco profile memory # core plus persistent recall and learning
  55. vibeco profile full # all agents, skills and rules; 53 of 74 hooks
  56. ```
  57. Profile changes reconcile installer-owned files and hook registrations. Modified or unmanaged files require review; the installer preserves ownership boundaries and creates backups before replacements. See [UPGRADING.md](UPGRADING.md).
  58. ## Token and runtime efficiency
  59. The default profile avoids installing the complete agent and skill catalog. Context-producing hooks use character limits: **4,000 per event and 12,000 per session** for lean profiles; **8,000/50,000** for `full`. These are limits on participating hook context, not limits on total model input, output, reasoning, tool results or billing.
  60. Token cost depends on the selected model, task, retrieved content and retries. Character counts and `estimated_tokens` are estimates; provider-reported usage is recorded separately when present. This project does not promise a fixed percentage or multiplier of token savings. Reproducible measurements belong with their fixtures and commands in [the verification record](docs/verification.md).
  61. The core skill entrypoint payload is smaller after focused rewrites: see the [reproducible byte comparison](docs/verification.md#core-skill-entrypoint-payload). Skills load on demand, so this is an available instruction size measurement, not a whole-session token claim.
  62. For routine work, stay on `core`, inspect relevant files, use one bounded implementation path and run the checks required for the change. Select `context`, `memory` or `orchestration` when the task benefits from them. Installing `full` does not automatically launch an agent swarm.
  63. ## How the runtime fits together
  64. ```mermaid
  65. flowchart LR
  66. P[Selected profile] --> I[Installer]
  67. I --> C[Selected agents and skills]
  68. I --> H[Registered Claude hooks]
  69. E[Claude lifecycle event] --> H
  70. H --> B[Bounded optional context]
  71. B --> W[Claude session]
  72. C --> W
  73. W --> V[Implementation and verification]
  74. ```
  75. Hooks run in response to host lifecycle events. They can inspect tool input, return a permission decision or provide context. They do not guarantee that a suggested agent or workflow will execute. Agent and skill files are instructions consumed by the host model.
  76. The optional memory pipeline captures and consolidates patterns under the user's Claude directory. Its thresholds are heuristics, not proof that a learned rule is correct. Review learned material before relying on it in sensitive work.
  77. ## Catalog
  78. | Path | Catalog (in repo) | Default core |
  79. | --- | --- | --- |
  80. | `agents/` | 138 Markdown agent definitions | 12 |
  81. | `skills/` | 296 top-level skill directories; some contain nested skills | 32 |
  82. | `hooks/src/` | 74 TypeScript entry points, including utilities and CLIs | 6 |
  83. | `rules/` | 20 rule documents; the lean profiles do not install the full rule set | - |
  84. The `full` profile installs 138 agents, 296 skills, 20 rules and 53 of the 74 hooks.
  85. Source entry points, unique hook names and registered commands are different counts. [hooks/hooks.json](hooks/hooks.json) is the registration source. Some files are command-line utilities or are not registered by default. Use `vibeco list agents --search security` to browse the installed catalog.
  86. ## Codex and other adapters
  87. | Host | Setup | Scope |
  88. | --- | --- | --- |
  89. | Claude Code | `./install.sh --profile core` | Selected agents, skills, hooks and rules |
  90. | Codex | `./install-codex.sh --profile core --install-luna-worker` | Selected skills and an optional bounded `luna_worker` |
  91. | Cursor | `./install-cursor.sh /path/to/project` | Project instructions, MDC rules and skills |
  92. | Other compatible hosts | Manual | Skill content; check the host's format and discovery rules |
  93. The Codex worker contract is [`.codex/agents/luna-worker.toml`](.codex/agents/luna-worker.toml): `gpt-5.6-luna` with `max` reasoning. The project config stays model-neutral. Claude-only `opus`/`sonnet` metadata is not applied by the Codex installer. Model availability and custom-agent support depend on the host and account. See [docs/codex-setup.md](docs/codex-setup.md).
  94. ## Optional integrations and data
  95. Core installation does not provision external services. Some optional hooks and skills need tools that are not bundled:
  96. - Code context and Python diagnostics use a compatible TLDR daemon; unavailable diagnostics can return no findings.
  97. - Memory graph features need a Node-compatible `better-sqlite3` installed in `~/.claude/hooks/`; a missing or incompatible module is reported and that graph operation skips. Other recall workflows reference Python/`uv` scripts and may need a configured database.
  98. - The `dashboard` profile adds one emitter to core (7 unique hook entry points across 8 registrations). It records agent starts on PreToolUse and structural tool completion events on PostToolUse; `full` includes it too. The dashboard uses the local `ws` package. Install it with `npm ci --ignore-scripts --prefix ~/.claude/tools/dashboard`, as described in [its setup guide](tools/dashboard/README.md).
  99. - Compiler assistance may contact a configured LM Studio endpoint. Braintrust extraction requires `VIBECO_BRAINTRUST_ANALYSIS=1`, a key and its script. GitHub-mutating hooks require `VIBECO_GITHUB_AUTOMATION=1`. MCP, search and provider skills may contact their configured services.
  100. Local state can contain code excerpts, prompts, tool output, errors and learned rules under `~/.claude/`. The core product is local; integrations may send data to the services you configure. Review the enabled profile, credentials and endpoint settings. The credential hook is a best-effort guard and does not replace host permissions or an operating-system sandbox. See [SECURITY.md](SECURITY.md).
  101. ## Development
  102. ```bash
  103. cd hooks
  104. npm ci
  105. npm run check
  106. npm run build
  107. npm test
  108. ```
  109. Keep `hooks/dist/` synchronized with source. Installer tests use temporary home directories; do not use a personal installation as a fixture. See [CONTRIBUTING.md](CONTRIBUTING.md), [ARCHITECTURE.md](ARCHITECTURE.md) and [docs/verification.md](docs/verification.md).
  110. ## Visuals
  111. The current [eight-second GIF](assets/runtime-demo.gif) is a locally rendered product diagram, not a screen recording or benchmark. Its [editable source](assets/runtime-demo-source/README.md) freezes data from the runtime manifest. The [static diagram](assets/runtime-overview.svg) describes profile selection and runtime boundaries. Earlier terminal GIFs remain in `assets/` as historical material; they depict legacy workflows and are not evidence that every feature runs in the default profile. See [assets/README.md](assets/README.md) for sources and reproduction requirements.
  112. <a name="turkce"></a>
  113. ## Türkçe
  114. vibecosystem, Claude Code için agent, skill ve hook'ları bir araya getirir. Varsayılan `core` profili, araştırma, geliştirme, inceleme ve doğrulama için seçilmiş bileşenleri kurar. Bellek, öğrenme ve çoklu agent akışları ayrıca seçilir; katalogdaki her şey her oturumda yüklenmez.
  115. ```bash
  116. ./install.sh --profile core --dry-run
  117. ./install.sh --profile core
  118. vibeco effective-config
  119. ```
  120. `vibeco effective-config` aktif profili ve kayıtlı hook sayısını gösterir.
  121. Token verimliliği için önce göreve uygun küçük profili seç. Bağlam sınırları modelin toplam token tüketimini sınırlamaz; ücret ve kullanım görevine, modele ve tekrar sayısına bağlıdır. Ölçülmemiş tasarruf oranları vaat etmiyoruz. `full` profili 138 agent, 296 skill ve 20 rule ile 74 hook'tan 53'ünü kurar; otomatik olarak bir agent ordusu başlatmaz.
  122. 3.5.1 şu anda npm üzerindeki güncel sürümdür. Yayınlanmış sürüm ve kurulum için İngilizce bölümü, değişiklikler için [CHANGELOG.md](CHANGELOG.md) dosyasını takip edebilirsin.
  123. ## Credits and license
  124. Built by [@vibeeval](https://x.com/vibeeval). MIT licensed. See [LICENSE](LICENSE).
  125. The catalog incorporates ideas credited in the individual skills and agent definitions, including work from Shannon, UI UX Pro Max, Game Studios, Skill Gateway, Pyxel, copilot-orchestra and Trail of Bits Skills. External tools and providers retain their own licenses and terms.
install-remote.sh · Bundled script · 119 lines · 1 finding marked
  1. #!/bin/bash
  2. # Usage: curl -fsSL https://raw.githubusercontent.com/vibeeval/vibecosystem/main/install-remote.sh | bash -s -- --profile core
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 2 on GitHub
  3. set -euo pipefail
  4. if [ -z "${HOME:-}" ] || [ "$HOME" = "/" ]; then
  5. echo "Error: HOME must be a non-root directory." >&2
  6. exit 1
  7. fi
  8. INSTALL_DIR="$HOME/.vibecosystem"
  9. REPO_URL="https://github.com/vibeeval/vibecosystem.git"
  10. INSTALL_ARGS=("$@")
  11. DRY_RUN=false
  12. REQUESTED_PROFILE=core
  13. usage() {
  14. echo 'Usage: install-remote.sh [--profile NAME] [--dry-run] [--force] [--prune] [--non-interactive]'
  15. }
  16. # Validate before cloning or updating. An explicit invocation never prompts.
  17. while [ "$#" -gt 0 ]; do
  18. case "$1" in
  19. --profile)
  20. [ "$#" -ge 2 ] || { echo 'Error: --profile requires a value.' >&2; exit 1; }
  21. REQUESTED_PROFILE="$2"
  22. shift
  23. ;;
  24. --profile=*) REQUESTED_PROFILE="${1#*=}" ;;
  25. --dry-run) DRY_RUN=true ;;
  26. --force|--prune|--non-interactive) ;;
  27. --help|-h) usage; exit 0 ;;
  28. *) echo "Error: unknown option: $1" >&2; exit 1 ;;
  29. esac
  30. shift
  31. done
  32. case "$REQUESTED_PROFILE" in
  33. core|quality|context|memory|orchestration|dashboard|full|frontend|backend|fullstack|devops|minimal|smart|all) ;;
  34. *) echo "Error: unknown profile: $REQUESTED_PROFILE" >&2; exit 1 ;;
  35. esac
  36. echo ''
  37. echo ' vibecosystem remote installer'
  38. echo ' =============================='
  39. echo ''
  40. command -v git >/dev/null 2>&1 || { echo 'Error: git is required.' >&2; exit 1; }
  41. command -v node >/dev/null 2>&1 || { echo 'Error: Node.js is required (https://nodejs.org/).' >&2; exit 1; }
  42. NODE_VER=$(node -v | sed 's/v//' | cut -d. -f1)
  43. if [ "$NODE_VER" -lt 18 ]; then
  44. echo "Error: Node.js >= 18 required (you have $(node -v))" >&2
  45. exit 1
  46. fi
  47. if [ -e "$INSTALL_DIR" ] || [ -L "$INSTALL_DIR" ]; then
  48. if [ -L "$INSTALL_DIR" ] || [ ! -d "$INSTALL_DIR/.git" ]; then
  49. echo "Error: $INSTALL_DIR is not a managed Git checkout. Preserving it." >&2
  50. exit 1
  51. fi
  52. origin="$(git -C "$INSTALL_DIR" remote get-url origin)"
  53. case "$origin" in
  54. "$REPO_URL"|"https://github.com/vibeeval/vibecosystem"|"git@github.com:vibeeval/vibecosystem.git") ;;
  55. *) echo "Error: unexpected origin at $INSTALL_DIR. Preserving it." >&2; exit 1 ;;
  56. esac
  57. if [ "$DRY_RUN" = false ]; then
  58. if [ -n "$(git -C "$INSTALL_DIR" status --porcelain)" ]; then
  59. echo "Error: $INSTALL_DIR has local changes. Commit or back them up before updating." >&2
  60. exit 1
  61. fi
  62. echo 'Updating existing installation...'
  63. if ! git -C "$INSTALL_DIR" pull --ff-only; then
  64. echo 'Error: update failed. Existing checkout and local files were preserved; retry when Git is ready.' >&2
  65. exit 1
  66. fi
  67. else
  68. echo 'dry-run: skipping Git updates; inspecting the existing checkout.'
  69. fi
  70. elif [ "$DRY_RUN" = true ]; then
  71. echo "dry-run: would clone $REPO_URL into $INSTALL_DIR and install profile $REQUESTED_PROFILE."
  72. echo 'No checkout or runtime files were changed.'
  73. exit 0
  74. else
  75. echo 'Cloning vibecosystem...'
  76. git clone --depth 1 "$REPO_URL" "$INSTALL_DIR"
  77. fi
  78. if [ "${#INSTALL_ARGS[@]}" -eq 0 ]; then
  79. echo ''
  80. echo 'Choose a profile:'
  81. echo ' 1) core Low-noise default runtime'
  82. echo ' 2) frontend React/Next.js focused'
  83. echo ' 3) backend API/DB/infra focused'
  84. echo ' 4) fullstack Frontend + backend'
  85. echo ' 5) devops CI/CD/cloud focused'
  86. echo ' 6) orchestration Explicit multi-agent workflows'
  87. echo ' 7) full Every legacy capability'
  88. if [ -t 0 ]; then read -r -p 'Select [1-7, default 1]: ' CHOICE; else CHOICE=1; fi
  89. case "$CHOICE" in
  90. 2) REQUESTED_PROFILE=frontend ;;
  91. 3) REQUESTED_PROFILE=backend ;;
  92. 4) REQUESTED_PROFILE=fullstack ;;
  93. 5) REQUESTED_PROFILE=devops ;;
  94. 6) REQUESTED_PROFILE=orchestration ;;
  95. 7) REQUESTED_PROFILE=full ;;
  96. *) REQUESTED_PROFILE=core ;;
  97. esac
  98. INSTALL_ARGS=(--profile "$REQUESTED_PROFILE" --prune --non-interactive)
  99. fi
  100. cd "$INSTALL_DIR"
  101. bash install.sh "${INSTALL_ARGS[@]}"
  102. if [ "$DRY_RUN" = true ]; then exit 0; fi
  103. echo ''
  104. echo 'Done! vibecosystem is ready.'
  105. echo ' vibeco help Show all commands'
  106. echo ' vibeco stats Ecosystem statistics'
  107. echo ' vibeco doctor Health check'
  108. echo ' vibeco dashboard Start optional monitoring UI'
  109. case ":$PATH:" in
  110. *":$HOME/.local/bin:"*) ;;
  111. *) echo 'Add ~/.local/bin to your shell PATH to run vibeco.' ;;
  112. esac

Version history

  • v3.5.111/12/2026

  • v3.5.011/5/2026

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.