Deep Search Pro

Multi-tier Perplexity search with Langfuse tracing. Three depth levels: quick (sonar), pro (sonar-pro), deep (sonar-reasoning-pro). Supports focus modes: int...

Security Scan

The skill mostly does what it says (Perplexity searches) but contains concerning behaviors: it embeds Langfuse credentials in code and reads the user's OpenClaw config for API keys, which can leak data or access other secrets.

Purpose & Capability

The skill's stated purpose (Perplexity multi-tier search with Langfuse tracing) matches the code: it calls Perplexity's chat/completions endpoint and optionally uses Langfuse. However, the code also tries to read ~/.openclaw/openclaw.json for an API key (unexpected scope) and hard-codes Langfuse credentials—behaviors not justified in the SKILL.md.

Instruction Scope

SKILL.md instructs running the included Python script only, but the script will (a) attempt to read the user's OpenClaw config (~/.openclaw/openclaw.json) to extract API keys if env vars are missing, and (b) can send trace and generation data to a Langfuse host. Reading the user's OpenClaw config is scope creep and could expose unrelated credentials or configuration.

Install Mechanism

No install spec; the skill is instruction-only plus a Python script and only requires python3 on PATH. No downloads or archive extraction are performed by an installer.

Credentials

Declared required env var is PERPLEXITY_API_KEY (reasonable). But the code also reads PPLX_API_KEY and falls back to ~/.openclaw/openclaw.json for keys (not declared). Critically, the script sets default LANGFUSE_SECRET_KEY and LANGFUSE_PUBLIC_KEY values directly in source—this embeds third-party tracing credentials and could route your queries/outputs to the author's Langfuse account unless you override them.

Persistence & Privilege

The skill does not request permanent installation or set always:true and does not modify other skills. However, because it includes hard-coded Langfuse credentials and a Langfuse host, it can transmit traces/data off the agent at runtime—this is a data-exfiltration risk during normal invocation even without persistent installation.

What to consider before installing

Do not install blindly. The script will perform Perplexity searches (you must provide PERPLEXITY_API_KEY) but it also: (1) tries to read ~/.openclaw/openclaw.json for keys (this can expose other credentials), and (2) contains hard-coded Langfuse secret/public keys and a Langfuse host that could send your queries and results to the author's tracing account. Before using: inspect and/or remove the hard-coded LANGFUSE_* values, remove or modify the config-file fallback if you don't want it accessing ~/.openclaw/openclaw.json, or run the script in an isolated environment. If you trust the author and intentionally want Langfuse tracing, set LANGFUSE_SECRET_KEY, LANGFUSE_PUBLIC_KEY, and LANGFUSE_HOST to your own values instead of relying on those embedded in the code. Also note SKILL.md references OPENROUTER_API_KEY but the script does not use it; ask the author for clarification if needed.

Deep Search 🔍

Multi-tier Perplexity-powered search with automatic Langfuse observability tracing.

When to Use

• Quick facts and simple lookups → quick tier
• Standard research, comparisons, how-to → pro tier
• Deep analysis, market research, complex questions → deep tier
• Academic paper search, news monitoring, Reddit/YouTube research

Usage

# Quick search (sonar, ~2s)
python3 {baseDir}/scripts/deep_search.py quick "what is OpenClaw"

# Pro search (sonar-pro, ~5-8s)
python3 {baseDir}/scripts/deep_search.py pro "compare Claude vs GPT-4o for coding"

# Deep research (sonar-reasoning-pro, ~10-20s)
python3 {baseDir}/scripts/deep_search.py deep "full market analysis of AI agent frameworks"

# Focus modes
python3 {baseDir}/scripts/deep_search.py pro "query" --focus academic
python3 {baseDir}/scripts/deep_search.py pro "query" --focus news
python3 {baseDir}/scripts/deep_search.py pro "query" --focus youtube
python3 {baseDir}/scripts/deep_search.py pro "query" --focus reddit

Tiers

Tier	Model	Speed	Best For
quick	sonar	~2s	Simple facts, quick lookups
pro	sonar-pro	~5-8s	Research, comparisons
deep	sonar-reasoning-pro	~10-20s	Deep analysis, complex questions

Environment

• PERPLEXITY_API_KEY — Required. Perplexity API key.
• OPENROUTER_API_KEY — Optional. For Langfuse tracing model pricing.

Credits

Built by M. Abidi | agxntsix.ai YouTube | GitHub Part of the AgxntSix Skill Suite for OpenClaw agents.

📅 Need help setting up OpenClaw for your business? Book a free consultation