newell-paul/sprint-scout
Plan cycling segment attempts around the wind. A browser-only planner and a showcase of AI-native engineering.
field: Data & analytics workflow: Testing & TDDDocsDebuggingPR / code reviewCI/CD & deploy runs claude in CI
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Protects secrets · Pre-tool screening hook · No YOLO mode · details
Quality 51/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit newell-paul/sprint-scout/.claude ./rig-sprint-scout # 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 newell-paul/sprint-scout/.claude/skills/bugfix .claude/skills/bugfix $ npx degit newell-paul/sprint-scout/.claude/skills/intent .claude/skills/intent $ npx degit newell-paul/sprint-scout/.claude/skills/release .claude/skills/release
$ curl -fsSL --create-dirs -o .claude/agents/comment-checker.md https://raw.githubusercontent.com/newell-paul/sprint-scout/main/.claude/agents/comment-checker.md $ curl -fsSL --create-dirs -o .claude/agents/test-runner.md https://raw.githubusercontent.com/newell-paul/sprint-scout/main/.claude/agents/test-runner.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/newell-paul/sprint-scout/main/.claude/agents/verifier.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Edit(./.env)",
"Edit(./.env.*)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Read|Edit|Write|MultiEdit|Bash",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard.mjs\""
}
]
}
]
}
} Skills (3)
Subagents (3)
| comment-checker model: sonnet | Checks that comments in tests and test helpers explain why, never what, using a fixed checklist so every run flags the same things. Use when asked to audit test comments. Never edits files. |
| test-runner model: haiku | Runs the project's test suite and reports only totals and failure output. Use whenever the user asks to run tests, the suite, or checks before a commit or PR. Do not run tests inline in Bash; delegate |
| verifier | Fresh-context final check before work is reported done. Reads the plan and the diff, runs the verification command, and returns pass or fail with evidence. Never fixes anything. |
Hooks (2)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Read|Edit|Write|MultiEdit|Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard.mjs" |
| PostToolUse | Edit|Write|MultiEdit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/after-edit.mjs" |
Permissions
deny (4)
Read(./.env)
Read(./.env.*)
Edit(./.env)
Edit(./.env.*)
ask (0)
—
allow (0)
—
Similar rigs
tomgould/mullet-public
Gated unattended coding jobs: Mullet runs a large coding task in a sandboxed clone and a step passes only on its own checks. Delegates to free, local, Kimi, OpenRouter or Claude models.
Minimalist 1.1k tok · · Q3
eugene1g/agent-safehouse
Sandbox your local AI agents so they can read/write only what they need
Minimalist 1.5k tok · · Q0
movieclaw/MovieClaw
Next-gen AI media server. One container replaces Sonarr, Radarr, Prowlarr, Bazarr, Jellyseerr and Jellyfin, with a web app, a native iPhone app, and an AI agent that gets things done. Free for personal use.
Minimalist 320 tok · · Q0
jgravelle/jdocmunch-mcp
The leading, most token-efficient MCP server for documentation exploration and retrieval via structured section indexing
Pragmatist 18.0k tok · · Q10