Skip to content
Back to skills

2528 Multi Hop Reasoning 2b924562

ASecurity

**Opus**: Chains 5+ reasoning steps naturally. Maintains coherence across long inference chains. **Haiku**: Reliable for 2-3 hops. Degrades at 4+. May skip intermediate steps or reach incorrect conclusions. **Mitigation**: Decompose into sequential sub-tasks with explicit outputs. ``` Step 1: Extract [A] from the input. Step 2: Using [A], determine [B] by [specific method]. Step 3: Given [B], select [C] from [enumerated options]. ``` Or bound reasoning depth: "Think in exactly 3 steps, writin...

  • 9 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 11, 2026
ai-agents

Security analysis

A100/100

Scanned October 11, 2026

npx -y skills add tools-only/X-Skills --skill 2528-multi-hop-reasoning_2b924562 --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of 2528 Multi Hop Reasoning 2b924562?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for 2528 Multi Hop Reasoning 2b924562
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tools-only-2528-multi-hop-reasoning-2b924562/badge)](https://www.skillsdirectory.com/skills/tools-only-2528-multi-hop-reasoning-2b924562)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

SKILL.md
# Multi-Hop Reasoning

**Opus**: Chains 5+ reasoning steps naturally. Maintains coherence
across long inference chains.

**Haiku**: Reliable for 2-3 hops. Degrades at 4+. May skip intermediate
steps or reach incorrect conclusions.

**Mitigation**: Decompose into sequential sub-tasks with explicit outputs.

```
Step 1: Extract [A] from the input.
Step 2: Using [A], determine [B] by [specific method].
Step 3: Given [B], select [C] from [enumerated options].
```

Or bound reasoning depth: "Think in exactly 3 steps, writing each step
before the next."

Never rely on Haiku to chain more than 3 inferences silently. If the
task requires 4+ hops, split into separate prompts or make intermediate
results explicit in the process steps.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.