Stats
Actions
Tags
From deepagents-builder
Run a maturity assessment on your agent architecture. Scores across 4 categories and identifies the path to the next level.
How this command is triggered — by the user, by Claude, or both
Slash command
/deepagents-builder:assess [agent-path]This command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# Assess Agent Maturity Run the 80-point maturity assessment on an agent architecture. ## Workflow ### Step 1: Locate Agent Code Find the agent to assess: 1. If `$ARGUMENTS` provides a path, use it directly 2. Otherwise, search for `create_agent` or `create_deep_agent` in: - `agent.py`, `main.py`, `src/**/*.py`, `agents/*/agent.py` 3. If multiple agents found, ask which to assess 4. If no agent found: Exit. ### Step 2: Trigger Evolution Guide Hand off to the `evolution-guide` agent in **assessment mode** with: - Agent file path - Any context about the project structure The...
Run the 80-point maturity assessment on an agent architecture.
Find the agent to assess:
$ARGUMENTS provides a path, use it directlycreate_agent or create_deep_agent in:
agent.py, main.py, src/**/*.py, agents/*/agent.pyNo agent code found.
→ Run /new-sdk-app to scaffold a new DeepAgents project.
Hand off to the evolution-guide agent in assessment mode with:
The evolution-guide will:
After the evolution-guide finishes, append:
Related commands:
/evolve — Guided refactoring to next level
/validate-agent — Detailed anti-pattern and security check
/tool-status — Tool catalog quality dashboard
/design-evals — Scaffold eval suite (key for Level 4)
npx claudepluginhub spulido99/claude-toolkit --plugin deepagents-builder/agent-native-auditAudits codebase against 8 agent-native architecture principles via parallel sub-agents, producing scored compliance report with gaps and recommendations.
/agent-native-auditAudits a codebase against 8 agent-native architecture principles, launching parallel sub-agents for scored compliance reports. Supports running a single principle when specified.
/audit-agentAudits Claude Code agent definitions for triggering accuracy, system prompt quality, tool appropriateness, and example coverage. Supports --full and --mini audit depths.
/primeAnalyzes codebase for agent readiness and proposes improvements. Use optional [path], --report-only for assessment without changes, or --fix-all to apply fixes automatically.
/assessRuns AI literacy assessment: scans repo for evidence and habitat docs, asks clarifying questions, produces timestamped document with fixes, recommendations, reflection, and README badge.
/assessAssesses a local repo against a wiki's research and market landscape, producing gap analysis, opportunities, and competitive insights.