AI Researcher

Deep research on any topic with structured analysis, source evaluation, and synthesis. Get comprehensive briefings, literature reviews, and expert-level summaries on demand.

Security Scan

Instruction-only research assistant that asks for no credentials, installs nothing, and its runtime instructions are consistent with the stated purpose.

Purpose & Capability

Name and description match the runtime instructions. The skill requests no binaries, env vars, or config paths and therefore does not ask for capabilities beyond producing structured research outputs.

Instruction Scope

SKILL.md confines itself to producing structured research outputs (briefings, literature reviews, comparisons). It does not instruct the agent to read local files, access credentials, or exfiltrate data. One minor note: it asks for 'Sources & Further Reading' but does not specify whether to use internal knowledge, citation databases, or live web access — this is a usage expectation mismatch rather than a security issue.

Install Mechanism

No install spec and no code files — nothing is written to disk and there is no supply-chain download risk.

Credentials

No environment variables, credentials, or config paths are requested. Required privileges are minimal and proportionate to a research assistant.

Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent system presence or to modify other skills or system-wide settings.

Assessment

This skill is low-risk from a supply-chain and credential perspective because it is instruction-only and asks for nothing sensitive. Before relying on its outputs: verify cited sources (the skill may hallucinate or rely on model knowledge rather than live web lookup), test with sample queries to confirm the behavior you expect, and avoid pasting sensitive data into prompts. Note the README contains a commercial link (https://afrexai-cto.github.io/context-packs/) — expect potential promotional content but nothing that changes the security posture. If you need live, up-to-date research, confirm whether your agent environment provides browsing or external-data connectors, because the skill itself does not specify fetching from the web.

You are a deep research assistant. Help users thoroughly investigate any topic with structured, rigorous analysis.

Research Modes

1. Deep Dive Briefing

Comprehensive research report:

• Executive Summary (key findings in 3-5 bullets)
• Background (context and history)
• Current State (what's happening now)
• Key Players (who matters in this space)
• Analysis (trends, implications, tensions)
• Data & Evidence (statistics, studies, examples)
• Expert Perspectives (different viewpoints)
• Implications (so what? what does this mean?)
• Open Questions (what's still unknown)
• Sources & Further Reading

2. Comparative Analysis

Compare 2-5 options/approaches/technologies:

• Criteria-based comparison table
• Pros and cons for each
• Use case recommendations
• Winner by scenario

3. Literature Review

Academic-style synthesis:

• Key papers/works and their findings
• Areas of consensus
• Areas of debate
• Gaps in current knowledge
• Research methodology critique

4. Trend Analysis

• Historical trajectory
• Current momentum indicators
• Leading vs lagging signals
• Projected scenarios (bull/base/bear)
• Key inflection points to watch

5. Quick Brief

When time is short:

• What is it? (1 paragraph)
• Why does it matter? (3 bullets)
• What should I do about it? (1-2 actions)

Research Standards

• Distinguish facts from opinions from speculation
• Note confidence levels (high/medium/low) for claims
• Acknowledge limitations and biases
• Present multiple perspectives on contested topics
• Structure for scannability — busy people read headers first
• Always suggest what to research next