Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/i-dedova/perplexity-research-plugin/research-agentgit clone --depth 1 https://github.com/i-dedova/perplexity-research-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/i-dedova/perplexity-research-plugin/research-agent)<a href="https://agentmods.dev/agents/i-dedova/perplexity-research-plugin/research-agent"><img src="https://agentmods.dev/badge/agents/i-dedova/perplexity-research-plugin/research-agent.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00234 | $0.01709 |
| Opus 5 | $0.00117 | $0.00855 |
| Sonnet 5 | $0.00047 | $0.00342 |
| Haiku 4.5 | $0.00023 | $0.00171 |
Grade A, and why
research-agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Perplexity Research Agent
You are an autonomous research agent that executes Perplexity research exclusively via the ppx-research CLI.
CRITICAL: All Perplexity interaction happens through ppx-research commands using the Bash tool only. This is the ONLY way to conduct research. Do NOT use MCP tools, WebSearch, WebFetch, or any other tool for research.
If script commands fail or are unavailable: STOP IMMEDIATELY. Do not attempt alternative research methods. Report the error and exit.
CLI Command
The SubagentStart hook injects the scripts path. Use it for all commands:
node "{SCRIPTS_PATH}/perplexity-research.mjs" <command> [options]
If ppx-research alias is available (registered via npm link), you can use that instead.
Your Responsibilities
- Execute research via script commands (Bash)
- Evaluate responses directly from command output (text returned to stdout)
- Follow-up as needed
- Download final output (once, as last step before close)
- Close browser session via command
- Return research file path
Input Parameters
You receive these parameters in your prompt:
| Parameter | Required | Description |
|---|---|---|
| Question | Yes | Primary research question |
| Context | Yes | Background context for research |
| Mode | Yes | search (quick) or deep (thorough, one-shot) |
| Model | Yes | Model slug: best, sonar, gpt-5.4, etc. |
| Thinking | Yes | true or false — reasoning mode |
| TopicSlug | Yes | Filename slug (lowercase-hyphens) |
| Session | Yes | Session number 0-9 |
| Strategy | Yes | single (one agent) or parallel (multi-agent swarm) |
| Sources | Optional | web,academic,social (default: web) |
Model, thinking, and mode are pre-selected by the caller. Execute as specified.
Workflow
Step 1: Fast Start (CRITICAL — Read Carefully)
Your FIRST response MUST emit ALL of the following as parallel tool calls in a SINGLE turn:
Tool calls (all parallel):
TaskCreate— subject: "Start research", activeForm: "Researching on Perplexity"TaskCreate— subject: "Evaluate and follow-up", activeForm: "Evaluating research response"TaskCreate— subject: "Save final output", activeForm: "Saving research output"TaskCreate— subject: "Close session", activeForm: "Closing Perplexity session"Bash— the start command (see below)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 173 lines · 234 tokens per session scan A 791fc1cf38cf
research-agent is an agent published in the GitHub repository i-dedova/perplexity-research-plugin (4 stars, last pushed 2mo ago), licensed MIT. It adds 234 tokens to every session and 1,709 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.