agentleFS
Sign inStart free
rezkamrezkam/boring-but-good/skills/perplexity/SKILL.md

perplexity

Skill1 starsChanged 14 days agoApache-2.0View on GitHub

Search, research, chat, and analyze files through Perplexity WebUI. Use when the user asks to use Perplexity, Perplexity WebUI, GPT thinking models through Perplexity, current-news research, cited web research, source-focused search, or file analysis.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the bundled scripts for all Perplexity work.

    line 9
  2. Distinctive:

    Run from the Perplexity skill directory:

    line 25
  3. Distinctive:

    Use direct model tools when the user asks for a specific tool capability by name.

    line 70
  4. Very specific:

    Use ask for normal agent work, current-news research, citations, source lists, file uploads, streaming, Spaces, and multi-turn turns in one …

    line 71

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Perplexity

Use the bundled scripts for all Perplexity work. Max-tier models are removed.

Setup

Install the Browser Tools dependency declared in package.json:

npm ci

Token safety

Never print or store the session token unless the user explicitly asks. Scripts read tokens from --token, --token-file, Browser Tools copied Chrome profile, PERPLEXITY_SESSION_TOKEN, or PPLX_SESSION_TOKEN. Browser Tools extraction starts a synced sandbox copy of the Default Chrome profile, navigates to Perplexity, and reads the session cookie through DevTools. Set PPLX_BROWSER_TOOLS_PROFILE or CHROME_PROFILE for another profile, or PPLX_BROWSER_TOOLS_TASK for a Browser Tools task profile. Commands retry transient WebUI 403 and query-processing failures according to --auth-retries. If an environment or token-file token is expired, commands retry once with Browser Tools when available. If token extraction fails but normal Chrome is logged in, verify the login is in the selected profile before trying another auth path.

Quick research

Run from the Perplexity skill directory:

uv run scripts/pplx.py ask "latest verified news about Iran war" \
  --model openai/gpt-5.4-thinking \
  --time-range week \
  --source-focus all \
  --citation-mode markdown \
  --format markdown

Direct model tools

Each direct tool is an executable script named for the tool. They all accept query, --search-focus, --source-focus, --time-range, --language, --latitude, --longitude, --format, and --save.

scripts/pplx_gpt54_thinking "research question" --source-focus all --time-range week

Available direct tools:

  • scripts/pplx_best
  • scripts/pplx_deep_research
  • scripts/pplx_sonar
  • scripts/pplx_gpt54
  • scripts/pplx_gpt54_thinking
  • scripts/pplx_gemini31_pro_think_low
  • scripts/pplx_gemini31_pro_think_high
  • scripts/pplx_claude_s46
  • scripts/pplx_claude_s46_think
  • scripts/pplx_kimi_k26_instant
  • scripts/pplx_kimi_k26_thinking
  • scripts/pplx_nemotron3_super_think

General commands

uv run scripts/pplx.py models                         # list exposed model IDs and direct tools
uv run scripts/pplx.py token                          # check token availability without printing it
uv run scripts/pplx.py token --validate               # live read-only auth check without printing it
uv run scripts/pplx.py ask "question"                 # direct client, sources, files, streaming, Spaces, multi-turn

When to use which path

  • Use direct model tools when the user asks for a specific tool capability by name.
  • Use ask for normal agent work, current-news research, citations, source lists, file uploads, streaming, Spaces, and multi-turn turns in one process.

More options and examples are in REFERENCE.md and EXAMPLES.md.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free