Upgrade Evaluator — Project Management agent for Claude Code
מעריך סקפטי ונפרד לתוצאות system-upgrader — קורא UPGRADE-PLAN/REPORT, מריץ detect.py, פותח את האפליקציה בדפדפן, בודק כל sprint contract בפועל, נותן ציון 1-10 לכל אחד מ-10 הממדים עם ראיות, ומחזיר PASS/.
How to install Upgrade Evaluator
Installs to ~/.claude/agents/ofirshudari1-ship-it-claude-tools-upgrade-evaluator.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/ofirshudari1-ship-it/claude-tools/HEAD/agents/upgrade-evaluator.md -o ~/.claude/agents/ofirshudari1-ship-it-claude-tools-upgrade-evaluator.md Restart Claude Code, or start a new session, for it to be picked up.
What Upgrade Evaluator does
name: upgrade-evaluator description: מעריך סקפטי ונפרד לתוצאות system-upgrader — קורא UPGRADE-PLAN/REPORT, מריץ detect.py, פותח את האפליקציה בדפדפן, בודק כל sprint contract בפועל, נותן ציון 1-10 לכל אחד מ-10 הממדים עם ראיות, ומחזיר PASS/FAIL ו-verdict. read-only על הקוד. השתמש אחרי כל ריצה של system-upgrader, או כשמבקשים - תעריך את השדרוג, תן ציון למערכת, בדוק אם זה באמת Done. tools: Read, Glob, Grep, Bash, mcp__Claude_Browser__*, mcp__playwright__* model: opus skills: [system-upgrade] maxTu
Alternatives in Project Management
- Agent Evaluator — Evaluates agent output against 5-axis quality rubric (accuracy, completeness, clarity, actionability, concisen 243.5k ★
- Deps Upgrade — You are a dependency management expert specializing in safe, incremental upgrades of project depende 31.9k ★
- Team Worker — Unified worker agent for team pipelines 530 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Evaluator
Behavioral active-verification agent. Phase 3 (active verifier: drives the running build against rubric criter
Delegate Pre Pass
Per-repo advisory delegation pre-pass for the /sprint --workflow Phase-5 review panel. Runs the gate + worktre
Grainulator
Autonomous research sprint subagent. Executes multi-pass research, adds claims, compiles, and generates output
Sprint Retro
Sprint retrospective agent — computes metrics, generates retrospective report with adaptive sizing recommendat
Ae Build
Agent Engineer build agent. Implements one plan step and loops against the project's deterministic gates until
Gan Contract Reviewer
GAN harness contract reviewer — audits a proposed sprint contract for specificity, comprehensiveness, scope, a
Related Skills
Conformance Walk
Conformance Walk — auto-detect the project flavor, run the matching CW adapter family (kailash reference engin
Sprint
Run the full engineering sprint: brainstorm -> plan -> build -> review -> qa -> polish-beta -> ship -> land-an
Gan Trinity
Three AI agents that check each other — Planner, Generator, Evaluator with contract negotiation. Multi-agent d