ronz204/deltas-ai
Living-contract documentation for AI coding agents: a verifiable contract between what your project says it does and what its code does.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit ronz204/deltas-ai/agents ./rig-deltas-ai/agents MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit ronz204/deltas-ai/agents/skills/auditor .claude/skills/auditor $ npx degit ronz204/deltas-ai/agents/skills/builder .claude/skills/builder $ npx degit ronz204/deltas-ai/agents/skills/comper .claude/skills/comper $ npx degit ronz204/deltas-ai/agents/skills/drafter .claude/skills/drafter $ npx degit ronz204/deltas-ai/agents/skills/scriber .claude/skills/scriber
$ curl -fsSL --create-dirs -o .claude/agents/specify.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/core/samples/specify.md $ curl -fsSL --create-dirs -o .claude/agents/approach.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/docs/approach.md $ curl -fsSL --create-dirs -o .claude/agents/expertise.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/docs/expertise.md $ curl -fsSL --create-dirs -o .claude/agents/overview.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/docs/overview.md $ curl -fsSL --create-dirs -o .claude/agents/structure.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/docs/structure.md $ curl -fsSL --create-dirs -o .claude/agents/notes.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/notes.md $ curl -fsSL --create-dirs -o .claude/agents/claims-fidelity.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/claims-fidelity.md $ curl -fsSL --create-dirs -o .claude/agents/code-change-scope.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/code-change-scope.md $ curl -fsSL --create-dirs -o .claude/agents/code-reliability.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/code-reliability.md $ curl -fsSL --create-dirs -o .claude/agents/commit-messages.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/commit-messages.md $ curl -fsSL --create-dirs -o .claude/agents/context-budgets.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/context-budgets.md $ curl -fsSL --create-dirs -o .claude/agents/done-needs-proof.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/done-needs-proof.md $ curl -fsSL --create-dirs -o .claude/agents/kb-deltas-routing.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/kb-deltas-routing.md $ curl -fsSL --create-dirs -o .claude/agents/project-language.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/project-language.md $ curl -fsSL --create-dirs -o .claude/agents/when-write-delta.md https://raw.githubusercontent.com/ronz204/deltas-ai/main/agents/rules/when-write-delta.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (5)
Subagents (15)
Similar rigs
gtback/dotfiles
My dotfiles
Minimalist 936 tok · · Q15
Stackbilt-dev/charter
Context compiler for AI coding agents. Modular .ai/ rules, deterministic routing, and generated configs for Claude, Codex, Cursor, and Gemini.
Pragmatist 341 tok · · Q13
celaldrg/subagentlint
Lint Claude Code subagent files (.claude/agents) before they fail silently: frontmatter typos, missing descriptions, duplicate names, broken hook scripts, description token budget. Zero-dependency Python CLI.
Orchestrator 247 tok · · Q0
BytesFromToby/plumbline
A "trust, but verify" workflow of AI-agent skills: prompt to verified code, every step trackable and deviations logged — plus an autonomous maintenance loop to keep a built project honest.
Pragmatist 1.7k tok · · Q0