# Assess Current Knowledge

> Map what's known vs assumed vs unknown

- Skill: `synapseradio/assess-current-knowledge` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add synapseradio/assess-current-knowledge`
- Raw SKILL.md: https://api.skillmd.com/api/skills/synapseradio/assess-current-knowledge/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: synapseradio (https://skillmd.com/u/synapseradio)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/synapseradio/assess-current-knowledge

---


Follow these steps:

### 1. Classify into four quadrants

- **Verified facts**: confirmed and directly supported
- **Known unknowns**: questions you know to ask — name each one
- **Unrecognized knowledge**: expertise or data you have but haven't applied here
- **Blind spots**: what you don't know you don't know — surface with probes, analogies, or outsider perspectives

### 2. Identify each item's source

- Direct observation
- Documentation
- Inference from related facts
- Assumption without evidence

Direct observation outweighs inference. Assumptions treated as facts propagate errors downstream.

### 3. Map the gaps

Make missing information and unverified assumptions explicit so they don't silently shape decisions. Some gaps close through investigation; others persist as acknowledged uncertainty.

### 4. Weigh readiness against risk

Compare the cost of being wrong against the effort of verification. High-consequence decisions demand higher confidence; low-stakes exploration can proceed with acknowledged uncertainty.

### 5. Close the critical gaps first

Prioritize information that would change decisions. Efficient sources:

- Configuration files for technical details
- Documentation for intended behavior
- Experiments for actual behavior

### 6. Recognize limits of competence

Know when to defer, acknowledge uncertainty, or mark claims as unverified. Some domains need expertise that isn't present; some questions can't be verified in the current context.