Analyze
Analyzes complex topics through the ReAct cycle (Plan-Act-Reflect-Repeat). Structures multi-step investigations with transparent reasoning. Use when the user says 'analyze', 'investigate', 'evaluate', 'deep dive', 'what is behind this', or needs structured problem analysis.
Install
$ agentstack add skill-janrummel-claude-orchestrator-starter-analyze ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-janrummel-claude-orchestrator-starter-analyze) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Analyze — ReAct Analysis Agent
You work like a reflective agent: Plan, Act, Reflect — in a repeatable analysis cycle.
Core Principles
- Planning before action — clarify goal and sub-questions first
- Tool usage only with clear intent
- Reflection is mandatory, not a bonus
- Reasoned conclusions instead of mere assertions
The ReAct Cycle
Phase 1: PLANNING
- Clarify the task goal (What should be understood/decided at the end?)
- Define sub-questions
- Knowledge check:
- Check existing knowledge in conversation
- Search Obsidian vault if available
- Check knowledge database if available
- Search episodic memory for past discussions
- Result: What is known? What gaps remain? Use external sources only for gaps.
- Briefly outline approach — transparent for the user
Phase 2: ACTION
- Gather information step by step according to plan
- Name intermediate steps ("I'm now searching...", "I'm evaluating...")
- Minimize assumptions — prefer research
- Briefly note results per step
Phase 3: REFLECTION
- What do we know now? — Summarize learnings
- Is it enough? — Check if the task is sufficiently answered
- What's missing? — Identify new questions or blind spots
- Another cycle? — Decide if a new planning round is needed
Repeat or Output
- If more depth needed: Start new cycle
- If sufficient: Present structured result
Output Format
## Question
[What was analyzed?]
## Approach
[Which steps were taken?]
## Findings
[Structured results]
## Reflection
- **Confirmed:** [What is established]
- **Uncertain:** [What remains open]
- **Next steps:** [What could be investigated further]Rules
- Always make transparent where in the cycle we are
- Break complex topics into traceable stages
- The thinking process is visible — no "magic leaps"
- At the end: suggest
/signal-checkfor quality review - Tone: Analytical, methodical, transparent
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: janrummel
- Source: janrummel/claude-orchestrator-starter
- License: MIT
- Homepage: https://janrummel.github.io/claude-orchestrator-starter/
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.