AGI Super Team

AGI Super Team is a versioned library of AI agent skills, specialist role packs, and human-in-the-loop team workflows for Codex and local coding-agent workspaces.

🧠 The system in one minute

The packs are designed for this review loop:

flowchart LR
  B["Brief"] --> C["Coordinator scopes"] --> S["Specialists execute with skills"] --> R["Reviewer challenges"] --> H["Human approves"]

AGI Super Team owns the versioned content, selection rules, safe copying, and repository checks. Your configured coding-agent harness owns the model, credentials, tools, execution, and final task output.

🎯 Start with an outcome

Need wider coverage? full-team selects all 14 manifest Agents. Start with a focused kit unless your evaluation genuinely needs every role.

⚡ Try it safely

Using Codex? Inspect the separately packaged Codex distribution. Its repository structure is tested; installation and loading in a current Codex client remain Validation pending.

To evaluate the generic workspace path, clone main and preview Solo Founder. Preview is the default and writes nothing:

git clone --depth 1 --branch main https://github.com/aAAaqwq/AGI-Super-Team.git
cd AGI-Super-Team
git rev-parse HEAD
./install.sh --source "$PWD" --destination /path/to/review-workspace solo-founder

Inspect the selected Agents and destinations. Apply only when they match your intent:

./install.sh --source "$PWD" --destination /path/to/review-workspace --apply solo-founder

The generic installer validates every required file before publishing staged workspaces. It preserves existing persona and skill files and rejects dangerous source or destination symlinks. See setup.md for prerequisites, updates, and recovery.

The manifest separates portable required/optional Skills from harnessSpecific catalog entries and unbundled external recommendations. Generic installs copy only classes that pass the current portability contract; the complete copied payload is scanned for known host paths and runtime-only commands.

Success: the preview writes nothing; apply creates three inspectable role workspaces without overwriting existing files.

🧭 Browse skills by outcome

The root README stays curated. Use the generated catalog when you need the complete, searchable inventory.

Explore the repository by depth:

🔎 Find high-quality skill sources

agent-skill-repository-index turns Daniel's reviewed source list into a safe selection workflow. Compare one candidate, inspect its permissions and provenance, then install or remove it without globally activating entire repositories.

Stars help with discovery, not trust. The matrix records DAILY, LIBRARY, and QUARANTINE boundaries; catalogs and runtimes are never bulk-installed.

✅ What is useful today

You can already browse a deterministic skill catalog, inspect every Agent instruction, preview a manifest-selected team, and assemble local role workspaces without overwriting existing files.

🧾 Reproducible installation receipt

Create a disposable destination, prove preview wrote nothing, apply, and verify the three expected Solo Founder workspaces:

AGI_SOLO_DEST="$(mktemp -d "${TMPDIR:-/tmp}/agi-solo-founder.XXXXXX")"

./install.sh --source "$PWD" --destination "$AGI_SOLO_DEST" solo-founder
test -z "$(find "$AGI_SOLO_DEST" -mindepth 1 -print -quit)"

./install.sh --source "$PWD" --destination "$AGI_SOLO_DEST" --apply solo-founder
test -f "$AGI_SOLO_DEST/workspace-ceo/SOUL.md"
test -f "$AGI_SOLO_DEST/workspace-pe/SOUL.md"
test -f "$AGI_SOLO_DEST/workspace-cco/SOUL.md"

python3 -m venv .venv
. .venv/bin/activate
python -m pip install --requirement requirements-dev.txt
npm test
npm run validate -- --warnings-as-errors
npm run check:skills
npm run check:taxonomy-evaluation
npm run check:architecture

This receipt proves manifest-driven selection, preview safety, staged copying, and repository integrity for the inspected checkout and destination state. It does not prove harness loading, task quality, or business outcomes.

The animation uses sanitized paths and is illustrative, not runtime evidence. Read the storyboard transcript.

Did the preview-first workflow help? Star the repository so you can find the verified path again.

🔌 Choose a distribution

The generic path requires Bash and Node.js; repository verification also requires npm and Python 3. Exact operating-system and client-version support remains bounded by CI and published receipts.

🗂️ Repository architecture

flowchart LR
  subgraph R["AGI Super Team repository: versioned content, not a runtime"]
    S["skills/<br/>reusable playbooks"]
    A["agents/<br/>14 role packs"]
    M["team-manifest.json<br/>4 kits + skill mappings"]
    C["plugins/agi-super-team-codex/<br/>curated Codex package"]
  end

  S --> I["install.sh<br/>preview → preflight → staged copy"]
  A --> I
  M --> I
  I --> W["workspace-agent<br/>inspectable local files"]
  W --> H["External harness<br/>model + tools + execution"]
  C --> H
  H --> O["Task artifacts<br/>behavioral evidence pending"]

  S --> G["Catalog generator"]
  M --> G
  G --> K["catalog/<br/>generated discovery index"]

  S --> V["Validator + tests"]
  A --> V
  M --> V
  I --> V
  V --> E["Repository receipt<br/>structure + install safety"]

For the authored-input, generated-output, distribution, and evidence boundaries, read the full repository architecture map, shared language, and decision records.

config/external-skill-sources.json records tombstones for removed machine-local links, including unresolved provenance fields. README text and generated catalog files are not inventory sources.

🧠 Team topology

Founder / operator
└── CEO: coordination and quality gates
    ├── CTO / PE: architecture and implementation
    ├── CPO / CCO / CMO: product, content, and growth
    ├── CQO / CFO / CDO: quantitative research, finance, and data
    ├── CLO / CRO / CSO / COO: legal, research, sales, and operations
    └── Governor: independent review and escalation

Mentor names are creative framing. They do not imply affiliation, endorsement, or guaranteed imitation.

🛡️ Boundaries and human approval

AGI Super Team is not a model, autonomous orchestrator, or agent runtime. Installing files does not make a harness load or execute them automatically.

• Inspect third-party commands and dependencies before execution.
• Never place credentials, private data, browser sessions, or production configuration in a skill or issue.
• Keep financial workflows in research or paper-trading environments until independently validated.
• Require explicit human approval for posts, messages, transactions, deployments, and destructive operations.
• Report vulnerabilities privately through GitHub Security Advisories.

🤝 Contribute and get help

• Report a reproducible issue
• Contributing and provenance
• Setup and recovery
• Security policy
• Growth playbooks
• MIT License

⭐ GitHub Stars

Track AGI Super Team's public growth over time. The live visualization is provided by Star History; click the chart to inspect the interactive timeline.