agentic-research

A local research agent that searches sources, checks its own answers, and provides citations. It can use a local document collection and runs through Ollama or another OpenAI-compatible service.

In plain words
What is it for?
Use it to research topics, search a local corpus, verify claims, and produce source-cited answers.
Why use it?
It makes the research process and verification decisions visible while avoiding a required cloud service.

Plugin for Claude Code

Install

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.

Claude Code
/plugin marketplace add TheAiSingularity/agentic-research-engine-oss
agentmods
npx agentmods add plugins/theaisingularity/agentic-research-engine-oss/claude_plugin
Clone the repo
git clone --depth 1 https://github.com/TheAiSingularity/agentic-research-engine-oss

Made for: Claude Code.

Per session not measured What this adds to a session before it is invoked.
When invoked not measured Not applicable: nothing here is loaded into a session.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

agentic-research 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 2d 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.

engine/mcp/claude_plugin/.claude-plugin/plugin.json · 42 lines

What it actually says

{
  "name": "agentic-research",
  "displayName": "Agentic Research — local, verified, open-source",
  "version": "0.1.3",
  "description": "Research agent that verifies its own answers with full source citations, optional local corpus, and zero cloud dependency. Runs on Gemma 3 4B via Ollama or any OpenAI-compatible endpoint. Every LLM call, source, and verification decision is visible. No telemetry.",
  "homepage": "https://github.com/TheAiSingularity/agentic-research-engine-oss",
  "repository": "https://github.com/TheAiSingularity/agentic-research-engine-oss",
  "author": {
    "name": "TheAiSingularity"
  },
  "license": "MIT",
  "keywords": [
    "research", "rag", "local", "ollama", "searxng", "verification",
    "cove", "citations", "agentic", "open-source", "mcp"
  ],
  "skills": [
    "skills/research.md",
    "skills/cite-sources.md",
    "skills/verify-claim.md",
    "skills/set-domain.md"
  ],
  "mcpServers": {
    "engine": {
      "command": "agentic-research-mcp",
      "args": [],
      "env": {
        "OPENAI_BASE_URL": "http://localhost:11434/v1",
        "OPENAI_API_KEY": "ollama",
        "MODEL_PLANNER": "gemma3:4b",
        "MODEL_SEARCHER": "gemma3:4b",
        "MODEL_SYNTHESIZER": "gemma3:4b",
        "MODEL_VERIFIER": "gemma3:4b",
        "MODEL_CRITIC": "gemma3:4b",
        "MODEL_ROUTER": "gemma3:4b",
        "MODEL_COMPRESSOR": "gemma3:4b",
        "EMBED_MODEL": "nomic-embed-text",
        "SEARXNG_URL": "http://localhost:8888"
      }
    }
  }
}
Changes

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.

  1. 2d ago First seen · 42 lines scan A 7da3c2ac9464

Subscribe to this mod's changes

agentic-research is a plugin published in the GitHub repository TheAiSingularity/agentic-research-engine-oss (1 stars, last pushed 4mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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.