Sb Judge — Development agent for Claude Code
Blind judge for strictlybetter.
How to install Sb Judge
Installs to ~/.claude/agents/nagisanzenin-strictlybetter-sb-judge.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/nagisanzenin/strictlybetter/HEAD/agents/sb-judge.md -o ~/.claude/agents/nagisanzenin-strictlybetter-sb-judge.md Restart Claude Code, or start a new session, for it to be picked up.
What Sb Judge does
name: sb-judge description: "Blind judge for strictlybetter. MUST BE USED by /strictlybetter:run for every promoted experiment before confirmation. Deliberately blind: receives ONE path to a payload holding the diff, the pre-registration, the screen numbers, the cards' gaming_risks, and a checklist; never the experimenter's reasoning or the conversation. Read-only; medium effort; returns ONLY a four-key verdict JSON." tools: Read effort: medium
You are strictlybetter's **blind judge**, t
Alternatives in Development
- Engram Assessor — Independent grader of learner productions for the Engram learning plugin 1.4k ★
- Latency Scanner — Scans codebase for request latency issues including sequential async operations, payload sizes, middleware ove 160 ★
- Mathodology Award Judge — Use for one independent blind judge seat scoring an award-level submission against contest rubric criteria 137 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Prompt Forager
Context-starved prompt forager for diversity injection. Receives ONE line (a purpose) and deliberately nothing
Blind Evaluator
Structurally separate eval agent. Receives ONLY the problem statement + rubric, NEVER the solution or the impl
Blind Reviewer
Blind read-only reviewer. Receives a complete
Discover Critic
Blind adversarial critic for a candidate product direction — decomposes it into ranked falsifiable assumptions
Drone Acceptance
Blind acceptance gate. Receives the human's brief, the repository and a run command - never plan.md, never sto
Phase7 Selfmod Auditor
Use proactively after every code change in the Phase-7 self-mod area (jarvis/core/self_mod/, pre-validate pipe
Related Skills
Oob
Out-of-band orchestrator — confirm BLIND SSRF/XXE/SQLi/RCE/Log4Shell by correlating interactsh callbacks to th
Design Bench
Run the ABC benchmark — reference URL → design-blind brief → render arms A/B/C → blind judge → scorecard with
Openprism Bakeoff
Blind-judge two models on one prompt to decide which earns a panel seat