Skillget

godmode

agent
Security Audit
Pass
Health Pass
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Community trust — 114 GitHub stars
Code Pass
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
  • Permissions — No dangerous permissions requested

No AI report is available for this listing yet.

SUMMARY

Engineering discipline for AI coding agents: question the request, verify against current docs, plan, build test-first on the cheapest capable model, review with a second model family, prove it before saying done. One repo for Claude Code, Codex, Antigravity, Cursor, Gemini CLI and OpenCode.

README.md

GodMode: engineering discipline for AI coding agents

Your coding agent questions the request, verifies against current docs instead of its memory, plans, builds test-first on the cheapest capable model, gets a second model family to review, and proves the result before it says done.

GitHub stars Version Validate 33 skills Zero dependencies Harnesses License

Install • Demo • Why • How it works • Skills • Harnesses • Compared with • Changelog

What a session looks like

A GodMode session: the agent questions the request, verifies Stripe's current API, plans seven tasks, runs them with a fresh subagent each and two reviews per task, gets a Codex second opinion, and proves the result with test, lint and build output before asking how to land it

An illustration of a session, not a transcript. Every line maps to a skill below.

Install

Claude Code

/plugin install godmode --marketplace NoobyGains/godmode

On older Claude Code versions use two steps: /plugin marketplace add NoobyGains/godmode, then /plugin install godmode@godmode. Start a new session afterwards: the bootstrap is injected at session start. /godmode:doctor confirms the install.

Antigravity

agy plugin install https://github.com/NoobyGains/godmode

Codex

codex plugin marketplace add NoobyGains/godmode
codex plugin add godmode@godmode

Any agent that reads skills (Cursor, Gemini CLI, OpenCode, Copilot CLI, Windsurf and others)

npx skills add NoobyGains/godmode

This installs the skills into the agent's skills directory. The bootstrap that makes them trigger automatically needs the harness-specific route in the harness table.

[!TIP]
Nothing to configure. GodMode is Markdown skills plus two small hooks (a session-start bootstrap and a one-line reminder per prompt); it never calls a network service, never collects telemetry, and has zero runtime dependencies.

Gemini CLI, Cursor, OpenCode, updating, uninstalling
  • Gemini CLI: gemini extensions install https://github.com/NoobyGains/godmode. The extension's GEMINI.md carries the bootstrap.
  • Cursor: install the plugin from this repository (the repo ships .cursor-plugin/plugin.json and .cursor-plugin/marketplace.json); a sessionStart hook injects the bootstrap.
  • OpenCode: add "godmode@git+https://github.com/NoobyGains/godmode.git" to the plugin (V1) or plugins (V2) array of opencode.json. Details in .opencode/INSTALL.md.
  • Update: Claude Code /plugin update godmode@godmode (third-party marketplaces do not auto-update); Antigravity: run the install command again; Codex: codex plugin marketplace upgrade then codex plugin add godmode@godmode.
  • Uninstall: Claude Code /plugin uninstall godmode@godmode; Antigravity agy plugin uninstall godmode; Codex codex plugin remove godmode.

Why GodMode

🔎 Grounded, not remembered

Model memory is a hypothesis. Before code is written against any API, CLI, library or platform rule, grounding verifies it against a current source and records the citation in the plan.

🧭 Should we build this at all?

rationale asks the senior engineer's first question before anyone designs: what problem, what effort, what already exists. Then intent-discovery turns the idea into an agreed design, one question at a time.

💸 Cost-disciplined orchestration

model-routing names the cheapest capable model and effort on every dispatch. using-godmode routes each task to the lightest execution mode that fits: inline, delegated, parallel, or an agent team only when teammates must coordinate.

✅ Evidence before "done"

completion-gate allows no success claim without fresh command output in the same message. code-review reviews every task twice: spec compliance, then quality.

🔁 A second model family

cross-model-review sends the diff to a reviewer from another model family (Codex, Gemini, or Claude) before landing. Different models make different mistakes.

🧩 One repo, every harness

Identical skills on Claude Code, Antigravity, Codex, Cursor, Gemini CLI and OpenCode. Skill bodies name actions; a per-harness reference maps them to real tools.

Without and with

Without GodMode the agent codes immediately from memory and declares success without running anything. With GodMode the same request is questioned, designed, grounded in current docs, planned, built test-first and proven.

How it works

GodMode is a gated pipeline. The agent cannot move to the next phase until the current gate has evidence.

The GodMode pipeline: define, plan, execute, review, ship. Each phase is a gate.

  1. Define. rationale challenges the request. intent-discovery asks one question at a time, presents two or three approaches, and writes the design document. grounding verifies every fact that can change.
  2. Plan. reference-first finds the proven implementation to build on. task-planning writes tasks of two to five minutes with exact files, complete code, commands and expected output.
  3. Execute. delegated-execution runs the plan with a fresh subagent per task, a ledger on disk that survives context loss, and a bounded fix loop; inline-execution is the cheap single-context mode; parallel-execution and team-orchestration cover independent and collaborating work. Every task is test-first.
  4. Review. code-review runs spec compliance then quality on each task; cross-model-review brings in another model family before landing; review-response handles feedback with evidence instead of reflexive agreement.
  5. Ship. quality-gates are machine-checked, completion-gate demands fresh output, merge-protocol lands the branch and cleans up the worktree.

Grounding: the skill nobody else makes the headline

The grounding skill refuses to code against a library from memory, fetches the current documentation, notes that the remembered API was outdated, and records the sources

Execution routing and model routing

GodMode picks an agent team because the modules must agree on conventions mid-work, sizes each seat with model-routing, runs three teammates in worktrees and merges them with the full suite green

Agent teams are the most expensive mode, so they are a decision, not a default: the lead enumerates the tasks, maps which ones need information from each other, and only forms a team when teammates must coordinate mid-work. On harnesses without teams the same skill degrades to parallel subagents.

The 33 skills

Every skill has a Prime Directive (the rule that cannot be broken), a Protocol, Cognitive Traps (the rationalisations it blocks) and Guardrails. Bodies describe actions, never one harness's tools; skills/using-godmode/references/ maps those actions to each harness.

Core loop
Skill Prime Directive
using-godmode Check for a relevant skill before any response or action
rationale No work without questioning whether the work is worth doing
intent-discovery No implementation without an agreed design
grounding No fact that can change is used without a current source
task-planning No implementation without a plan of atomic, verifiable tasks
knowledge-capture No meaningful task ends without recording what it taught
Execution
Skill Prime Directive
delegated-execution Fresh subagent per task, two reviews per task, a ledger that survives
inline-execution One context, every task, one final review
parallel-execution No concurrent dispatch without confirmed isolation
team-orchestration No team without a collaboration requirement
agent-briefing No dispatch without a structured brief
workspace-isolation No feature work on the main branch
model-routing No dispatch without a named model and effort, the cheapest that can do the job
merge-protocol No integration without passing tests
Quality
Skill Prime Directive
test-first No production code without a failing test first
code-review No landing without spec review, then quality review
cross-model-review No landing without one review from another model family
quality-gates No code lands with a failing machine check
completion-gate No completion claim without fresh verification output
review-response Every piece of feedback gets a technical evaluation
fault-diagnosis No fix without the root cause
error-recovery No further attempt without acknowledging the failure count
Research and design
Skill Prime Directive
reference-first No building without a proven reference
codebase-conventions Every addition mirrors an existing precedent
ui-research No UI without a researched pattern and design tokens
ui-engineering No component without structure, states and accessibility defined
system-design No structural complexity without a current requirement
Platform and meta
Skill Prime Directive
environment-awareness No shell command without knowing the target environment
project-bootstrap No feature code before the project foundation exists
security-protocol No external data reaches a system call, query or output unvalidated
performance-tuning No optimisation without a measurement proving the problem
writing-skills Skills are written test-first, like code
doctor Diagnose the install before blaming the skills

Harnesses

Harness Install Bootstrap at session start Status
Claude Code one-step /plugin install line above SessionStart hook, plus a one-line reminder on each prompt Verified on 2.1.310: hook tested on sh, with an empty PATH and through cmd.exe; in a clean profile (only Claude Code's bundled plugins present) a headless "Let's make a react todo list" invoked rationale first and answered with options, no code (evidence)
Antigravity CLI (agy) agy plugin install <repo url> Always-on plugin rule plus the skill index Verified on agy 1.4.7 (Windows): installed as a native plugin with all 33 skills; a headless "Let's make a react todo list" on Gemini Flash answered with the rationale analysis, no code (evidence)
Codex CLI and App codex plugin marketplace add + codex plugin add Native skill discovery ($godmode:using-godmode) Verified on Codex CLI 0.163: marketplace registered from this repo and godmode@godmode 2.0.0 installed and enabled (evidence)
Gemini CLI gemini extensions install <repo url> Extension context file Manifest shipped, untested
Cursor plugin from repository sessionStart hook Manifest shipped, untested
OpenCode V1 and V2 opencode.json plugin entry V1: injected into the first user message; V2: session context hook Plugin shipped and module-loads under Node; untested on a live install
Copilot CLI, Windsurf, others npx skills add NoobyGains/godmode None (skills trigger from their descriptions) Skills only

"Untested" means no maintainer has run it on a live install yet. If you do, open an issue with the transcript and the row changes. The per-prompt reminder is a UserPromptSubmit hook, which fires in interactive sessions; in our headless claude -p runs it did not fire, so the session-start bootstrap alone produced the verified behaviour.

Limits worth knowing: agent teams exist only on Claude Code and are experimental and opt-in there; Antigravity has no todo tool, so plans are tracked in a task artifact; OpenCode V2 has no todo tool either; Codex needs multi-agent enabled for subagent dispatch. Each references/<harness>-tools.md spells out the substitutions.

Compared with

GodMode grew out of superpowers and shares its shape: Markdown skills, a bootstrap at session start, a define-plan-execute-review-ship flow. The differences are the three commitments above, and this table is an attempt to be fair about when the others are the better pick.

Project What it is Use it instead when
superpowers 15 tightly tuned skills, 17 harnesses, an eval lab and a Discord You want the smallest core with the longest track record
compound-engineering Thirty-six skills with a learning loop that writes solutions back into the repo You want every fix to compound into a project knowledge base
spec-kit A CLI-driven spec-first workflow with verdict states Your team wants formal specifications as the primary artefact
agent-skills 25 lifecycle skills installed with npx skills add You want a lighter set without a bootstrap hook
GodMode 33 skills, 6 harness manifests, web-grounded facts, per-seat model routing, cross-model review You want the agent to verify instead of remember, spend tokens deliberately, and prove results

Evals

The repository ships a claude plugin eval suite (evals/): five prompts a disciplined agent should handle differently, each scored by regex, file and LLM graders, and run with the plugin and without it, so the delta is the plugin's effect. Numbers below are the mean of three runs per arm.

Case With GodMode Without Delta Note
design-before-code ("Let's make a react todo list") 1.00 0.67 +0.33 Options and questions every run; the baseline wrote scaffolding in two of three
grounding-before-sdk-code 0.67 0.17 +0.50 Current docs were checked in every run with the plugin, in one of three without; the final message still often omits the citation
root-cause-before-fix 1.00 0.83 +0.17 The baseline is already careful here
no-done-without-evidence 1.00 1.00 0 The baseline already refuses a bare "done"; kept as a regression guard
execution-mode-routing 0.50 0.50 0 A mode is named in every run, but not a model per seat; the open gap in model-routing
Mean 0.83 0.63 +0.20

Measured on Claude Code 2.1.310 with sonnet as the agent model and haiku as the judge on 2026-10-25, cost $2.97 for the thirty sessions; raw results in docs/verification/. Run it yourself:

claude plugin eval . --trust-plugin --runs 3 --model sonnet --judge-model haiku --no-publish --max-cost-usd 12

Verified before every release

  • node scripts/validate.js --ci: frontmatter rules, section order, cross-references, tool-name coupling, hook quoting and executable bits, version agreement across every manifest, README skill coverage, icon size.
  • bash tests/test-hooks.sh on Linux, macOS and Windows Git Bash, and tests/test-hooks-windows.ps1 through cmd.exe.
  • claude plugin validate . and agy plugin validate ..
  • A headless acceptance session per verified harness: the prompt "Let's make a react todo list" must trigger rationale or intent-discovery before any code.
  • One review of the release diff by a different model family than the author.

What GodMode is not

  • Not a code generator. It governs how the agent works, not what it builds.
  • Not magic. Composable discipline in Markdown; read any skill and you know exactly what it asks for.
  • Not free of judgement. Skills decide what evidence is needed; you still decide what to build.
  • Not locked in. MIT. Every skill is a standalone file. Fork it, trim it, or run npx skills add and take only the ones you want.
  • Not a telemetry source. No network calls, no analytics, no phone-home of any kind.

Contributing

Issues and pull requests are welcome. Read AGENTS.md first: it has the rules for skill wording (actions, not tools; facts from current sources; the bootstrap size limit) and the checks CI runs. New skills follow writing-skills.

Star history of NoobyGains/godmode

Thanks

  • @ivnad95, @g7trader and @AdversaLLC for diagnosing the 1.0 hook bug with exact reproductions; your fix is in 2.0.
  • @urbabby, whose orchestrator-core fork skill shaped the "rulings, not stalls" idea in delegated-execution.
  • Jesse Vincent and the superpowers contributors for the methodology GodMode grew from, and for the cross-harness hook and manifest patterns (MIT) that 2.0 follows.

License

MIT. See LICENSE.

Reviews (0)

No results found