By 89jobrien
Rust-native development methodology: TDD, debugging, collaboration patterns, and proven techniques for Rust workspaces
- Read the full error output before proposing any fix. Do not skim.
- Always run `git branch --show-current` before any commit. If on main, STOP.
- Cap parallel subagents at 5 concurrent.
- Default to read-only. Do not modify files unless the command explicitly
- Plan files go in `.ctx/tasks/`, not `.ctx/plans/`.
Dialectic proposer — Advocate stance. Argues for the most straightforward, favorable interpretation of a question. Used as a proposer in the dialectic synthesis pipeline.
Dialectic proposer — Alternative stance. Proposes a fundamentally different approach that reframes the question or solves it by other means entirely. Used as a proposer in the dialectic synthesis pipeline.
Dialectic pipeline orchestrator. Frames a question into adversarial roles, dispatches dialectic-advocate, dialectic-skeptic, and dialectic-alternative in parallel, then passes proposals to dialectic-synthesizer. Use when a decision benefits from structured adversarial analysis before committing to an approach.
Dialectic proposer — Skeptic stance. Finds failure modes, edge cases, and reasons to reject or challenge the premise. Used as a proposer in the dialectic synthesis pipeline.
Dialectic synthesizer. Reads all proposer outputs, maps disagreements, and produces a single reconciled answer with rationale, open questions, and rejected positions.
Patterns and techniques for adding governance, safety, and trust controls to AI agent systems in Rust. Use this skill when building agents that call external tools, implementing policy-based access controls, adding semantic intent classification to detect dangerous prompts, creating trust scoring systems for multi-agent workflows, building audit trails, or enforcing rate limits and content filters. Covers policy composition, tool wrappers, trust decay, and JSONL audit trails — all in Rust.
Use when creating a new local skill or fixing an existing skill that was saved to the wrong skills tree. Symptoms - a skill was written under ~/.claude/skills, the active setup uses ~/.agents/skills, or skill instructions still reference the old path convention.
Use when adding new BAML types/functions to cruxx-agentic. Symptoms - need to add structured extraction functions, mirror types to cruxx-types.
Use before any creative work — new features, new components, architecture decisions, adding crates, designing APIs. Triggers on "let's build", "I want to add", "design X", "how should we structure", or any request that would produce new code structure.
Commit and push workflow with pre-flight validation and hook recovery. Use when the user says "cap", "commit and push", or "ship it". Runs cargo nextest, fmt, clippy, stages all changes, writes a conventional commit, handles pre-commit hook failures, and pushes.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Self-contained Rust-native development methodology plugin for Claude Code.
Inspired by and built on the ideas in superpowers by obra — the original agentic skills framework for Claude Code. Godmode replaces the superpowers runtime with a Rust-backed CLI and task graph, but the methodology, skill structure, and session ritual are directly descended from that work.
Godmode combines a Claude Code skill set with a CLI-backed task graph. The binary owns all
stateful operations — skills are thin wrappers that call godmode and act on the output.
Tasks persist in .ctx/godmode/tasks.yaml (gitignored) across sessions via causal
depends_on chains.
cargo build --release -p godmode-cli && cp target/release/godmode ~/.cargo/bin/godmode
claude plugin install godmode@bazaar
Install the godmode pre-commit hook into any git repo to block commits when tasks are still running and to run cargo gates (fmt-check, clippy, nextest):
# From the repo root where you want the hook installed:
nu /path/to/godmode/hooks/install.nu
# Or, if godmode is installed as a plugin and $CLAUDE_PLUGIN_ROOT is set:
nu "$CLAUDE_PLUGIN_ROOT/hooks/install.nu"
The hook:
godmode handoff --json — exits non-zero if any task is in running state,
printing the offending task IDs.cargo fmt --all --check, cargo clippy --workspace -- -D warnings, and
cargo nextest run --workspace.If godmode is not on PATH, the task-state check is skipped gracefully and only the
cargo gates run.
| Skill | When |
|---|---|
godmode:using-godmode | Session orientation, available skills, rules |
godmode:task-driven-development | Implementing any feature or fix (TDD + YAML task list) |
godmode:systematic-debugging | Any bug, test failure, unexpected behavior |
godmode:brainstorm | Before any creative or design work |
godmode:writing-plans | Multi-step task with a spec or requirements |
godmode:verification-before-completion | Before claiming work is done |
godmode:task-management | Creating, tracking, executing a task graph |
godmode:parallel-agents | 2+ independent tasks to run concurrently |
godmode:code-review | Quality pass before merge |
godmode:refactoring | Restructure code without changing behaviour |
godmode:receiving-review | Process incoming review feedback |
godmode:cap | Commit and push with validation |
godmode:ci-fix | Fix a failing CI pipeline |
godmode:tackle-issues | Work GitHub issues in parallel worktrees |
godmode:testing-philosophy | Choose the right test type for the situation |
godmode:introspection | Audit skills and plugin files for consistency |
godmode:observability-as-infrastructure | Query and tail the session trace log |
godmode:wave-integration | Merge parallel agent branches into one commit |
godmode:moa | Multi-model reasoning via mixture of agents |
godmode:todo-issue-sync | Audit inline TODOs and sync to GitHub issues |
godmode:self-reflect | Session retrospective — patterns and surprises |
godmode:decompose | Break a large diff/PR into smaller independent PRs |
godmode:merge | Merge branches, resolve conflicts, create PRs |
godmode:agent-governance | Governance and trust controls for AI agent systems |
godmode:context-map | Map all files relevant to a task before changes |
godmode:doublecheck | Three-layer verification of AI-generated output |
godmode:rust-conventions | Rust coding conventions and best practices |
godmode:mini-context-graph | Persistent knowledge graph for codebase exploration |
godmode:memory-banking | Generate and maintain .ctx/memory-bank/ context |
npx claudepluginhub 89jobrien/bazaar --plugin godmodeAbsorb foreign repos into the Rust ecosystem — extract, classify, and generate hexagonal components
Session handoff toolkit — start, end, and visualize project state via HANDOFF.yaml and .ctx/.
Task and todo management specialist — doob CLI integration, council report ingestion, HANDOFF reconciliation
1Password and direnv session management — secrets validation and env chain tracing
Personal dev workflow plugin — Rust gates, code review, CI, git safety, multi-repo pulse
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Complete developer toolkit for Claude Code
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams