analyze-agent
Systematically analyze an AI agent's source code across 8 core + 4 optional platform dimensions and output structured analysis to docs/<agent>.md
npx skills add majiayu000/claude-skill-registry --skill analyze-agent --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Analyze Agent Deep-dive analysis of an AI agent's implementation. Reads the agent's source code from `projects/<agent>/` and fills `docs/<agent>.md` using the 8+4 dimension template. ## Trigger `/analyze-agent <agent-name>` ## Prerequisites 1. The agent's source code must exist at `projects/<agent-name>/`（run `./scripts/manage-submodules.sh add <agent-name>` if not present） 2. The analysis doc `docs/<agent-name>.md` should exist（run `./scripts/new-analysis.sh <agent-name>` if not present） ## Analysis Framework Analyze the agent across 8 core dimensions (always), plus 4 optional platform dimensions (when applicable). ### Platform Detection Before starting analysis, check if this is a platform-level agent: - Read `agents.yaml` for `type: agent-platform` field - Or detect signals: multi-channel support, gateway architecture, scheduling, persistent memory, companion apps - If platform agent: fill D1-D12 - If pure coding agent: fill D1-D8 only, delete or mark D9-D12 as "不适用" ### Core Agent Reference If the platform agent's core coding engine is another agent already analyzed (e.g., OpenClaw uses pi-agent): - D2-D6 should **reference** the existing analysis (link to `docs/<core-agent>.
- Trigger
- Prerequisites
- Analysis Framework
- Platform Detection
- Core Agent Reference
- Dimension 1: Overview & Architecture
- Dimension 2: Agent Loop
- Dimension 3: Tool/Action System
- Dimension 4: Prompt Engineering
- Dimension 5: Context Management
- Dimension 6: Error Handling & Recovery
- Dimension 7: Key Innovations
- Dimension 7.5: MVP Component Map
- Dimension 8: Cross-Agent Comparison
What does the analyze-agent skill do?
Systematically analyze an AI agent's source code across 8 core + 4 optional platform dimensions and output structured analysis to docs/<agent>.md
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill analyze-agent --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.