qf-studio/navigator
Finish What You Start — Context engineering for Claude Code. Sessions last 20+ exchanges instead of crashing at 7.
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 qf-studio/navigator/.claude ./rig-navigator # inspect, then merge into .claude/ 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 qf-studio/navigator/skills/backend-endpoint .claude/skills/backend-endpoint $ npx degit qf-studio/navigator/skills/backend-test .claude/skills/backend-test $ npx degit qf-studio/navigator/skills/database-migration .claude/skills/database-migration $ npx degit qf-studio/navigator/skills/frontend-component .claude/skills/frontend-component $ npx degit qf-studio/navigator/skills/frontend-test .claude/skills/frontend-test $ npx degit qf-studio/navigator/skills/nav-brief .claude/skills/nav-brief $ npx degit qf-studio/navigator/skills/nav-compact .claude/skills/nav-compact $ npx degit qf-studio/navigator/skills/nav-deep-research .claude/skills/nav-deep-research $ npx degit qf-studio/navigator/skills/nav-diagnose .claude/skills/nav-diagnose $ npx degit qf-studio/navigator/skills/nav-features .claude/skills/nav-features $ npx degit qf-studio/navigator/skills/nav-graph .claude/skills/nav-graph $ npx degit qf-studio/navigator/skills/nav-init .claude/skills/nav-init $ npx degit qf-studio/navigator/skills/nav-loop .claude/skills/nav-loop $ npx degit qf-studio/navigator/skills/nav-marker .claude/skills/nav-marker $ npx degit qf-studio/navigator/skills/nav-onboard .claude/skills/nav-onboard $ npx degit qf-studio/navigator/skills/nav-pilot .claude/skills/nav-pilot $ npx degit qf-studio/navigator/skills/nav-profile .claude/skills/nav-profile $ npx degit qf-studio/navigator/skills/nav-release .claude/skills/nav-release $ npx degit qf-studio/navigator/skills/nav-simplify .claude/skills/nav-simplify $ npx degit qf-studio/navigator/skills/nav-skill-creator .claude/skills/nav-skill-creator $ npx degit qf-studio/navigator/skills/nav-sop .claude/skills/nav-sop $ npx degit qf-studio/navigator/skills/nav-start .claude/skills/nav-start $ npx degit qf-studio/navigator/skills/nav-stats .claude/skills/nav-stats $ npx degit qf-studio/navigator/skills/nav-sync-claude .claude/skills/nav-sync-claude $ npx degit qf-studio/navigator/skills/nav-task .claude/skills/nav-task $ npx degit qf-studio/navigator/skills/nav-triz .claude/skills/nav-triz $ npx degit qf-studio/navigator/skills/nav-upgrade .claude/skills/nav-upgrade $ npx degit qf-studio/navigator/skills/nav-workflow .claude/skills/nav-workflow $ npx degit qf-studio/navigator/skills/plugin-slash-command .claude/skills/plugin-slash-command $ npx degit qf-studio/navigator/skills/product-design .claude/skills/product-design $ npx degit qf-studio/navigator/skills/visual-regression .claude/skills/visual-regression
$ curl -fsSL --create-dirs -o .claude/agents/deep-research-critic.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/deep-research-critic.md $ curl -fsSL --create-dirs -o .claude/agents/deep-research-fetcher.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/deep-research-fetcher.md $ curl -fsSL --create-dirs -o .claude/agents/deep-research-patcher.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/deep-research-patcher.md $ curl -fsSL --create-dirs -o .claude/agents/deep-research-writer.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/deep-research-writer.md $ curl -fsSL --create-dirs -o .claude/agents/navigator-research.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/navigator-research.md $ curl -fsSL --create-dirs -o .claude/agents/task-planner.md https://raw.githubusercontent.com/qf-studio/navigator/main/agents/task-planner.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (31)
backend-endpointbackend-testdatabase-migrationfrontend-componentfrontend-testnav-briefnav-compactnav-deep-researchnav-diagnosenav-featuresnav-graphnav-initnav-loopnav-markernav-onboardnav-pilotnav-profilenav-releasenav-simplifynav-skill-creatornav-sopnav-startnav-statsnav-sync-claudenav-tasknav-triznav-upgradenav-workflowplugin-slash-commandproduct-designvisual-regression
Subagents (6)
| deep-research-critic model: opus | Adversarial reader of a nav-deep-research draft. Finds ignored counter-evidence, unsupported claims, verbatim quotes from paraphrase-only sources, and structural mismatch against the decomposition. Wr |
| deep-research-fetcher model: sonnet | Fetches one disjoint batch of URLs into a nav-deep-research run as fenced source notes. Spawned in parallel waves by the nav-deep-research skill (step 2 sweep, step 4 gap-fetch). Never searches for ne |
| deep-research-patcher model: opus | Applies critic findings to a nav-deep-research draft as surgical Edit hunks and records a patch log. Tool-locked so it cannot regenerate the report. Spawned once at step 5 by the nav-deep-research ski |
| deep-research-writer model: opus | Writes the single report.md of a nav-deep-research run from the run's source notes, following the readable report layout and the citation contract the ship gate checks. Spawned once by the nav-deep-re |
| navigator-research model: sonnet | Specialized codebase exploration and architecture discovery. Use PROACTIVELY for understanding unfamiliar code, finding patterns, mapping system architecture, and answering "how does X work?" question |
| task-planner model: sonnet | Implementation planning and task breakdown specialist. Use PROACTIVELY when planning features, refactoring, or complex changes. Creates detailed implementation plans with effort estimates. |
Similar rigs
PedroMosquera/squadai
One config. Every AI agent. Zero drift. Idempotent setup for OpenCode, Claude Code, Cursor, Windsurf, and Copilot.
Fort Knox 2.7k tok ·
aaronsb/claude-code-config
Event-driven policy, process, and governance for Claude Code. Ways encode how we do things — prescriptive rules triggered by context, not requested by intent.
Orchestrator 1.3k tok ·
NeamulMorshed/The-Monkey-Brain
—
Skill Collector 11.6k tok ·
mblode/agent-skills
Nobody ships AI slop on purpose. These skills make sure you don’t.
Skill Collector 3.4k tok ·