Narrativediff

pip install cognis-narrativediff
narrativediff scan .            # → prioritized findings in seconds

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ narrativediff-emit --version
narrativediff 0.1.0

$ narrativediff-emit --help
usage: narrativediff [-h] [--version] [--format {table,json}]
                     {diff,outlets} ...

News bias & framing diff across many outlets per event.

positional arguments:
  {diff,outlets}
    diff                full bias/framing diff of a corpus JSON
    outlets             quick per-outlet bias one-liners

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  --format {table,json}
                        output format (default: table)

Blocks above are real narrativediff output — reproduce them from a clone.

Sample result format (illustrative values — run on your own data for real findings):

{
  "Findings": [
    {
      "id": "1234567890",
      "title": "Suspicious Network Traffic",
      "description": "Network traffic detected from an unknown IP address",
      "severity": "medium",
      "created_at": "2023-02-15T14:30:00Z"
    },
    {
      "id": "2345678901",
      "title": "Malware Detection",
      "description": "A suspicious executable was detected on the system",
      "severity": "high",
      "created_at": "2023-02-16T10:45:00Z"
    }
  ]
}

Usage — step by step

narrativediff diffs news bias and framing across many outlets for a single event. Console script: narrativediff.

1. Install from a clone:

   pip install -e .
   

2. Run the full bias/framing diff on an event corpus JSON:

   narrativediff diff corpus.json
   

3. Quick per-outlet scan — one bias line per outlet:

   narrativediff outlets corpus.json
   

4. Read the output — --format json for downstream analysis:

   narrativediff --format json diff corpus.json | jq '.bias_spread, .divergence_ranking'
   

5. Automate — batch many events through a pipeline:

   for f in events/*.json; do narrativediff --format json diff "$f" > "out/$(basename "$f")"; done
   

Contents

• Why narrativediff? · Features · Quick start · Example · Architecture · AI stack · How it compares · Integrations · Install anywhere · Related · Contributing

Why narrativediff?

News bias & framing diff across 50+ outlets per event — without standing up heavyweight infrastructure.

narrativediff is single-purpose, scriptable, and self-hostable: point it at a target, get prioritized results in the format your workflow already speaks (table · JSON · SARIF), gate CI on it, and let agents drive it over MCP.

Features

• ✅ Analyze Event
• ✅ Load Corpus
• ✅ Result To Dict
• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

pip install cognis-narrativediff
narrativediff --version
narrativediff scan .                       # scan current project
narrativediff scan . --format json         # machine-readable
narrativediff scan . --fail-on high        # CI gate (non-zero exit)

Example

$ narrativediff scan .
  [HIGH    ] NAR-001  example finding             (./src/app.py)
  [MEDIUM  ] NAR-002  another signal              (./config.yaml)

  2 findings · risk score 5 · 38ms

Architecture

flowchart LR
  IN[input] --> P[narrativediff<br/>analyze + score]
  P --> OUT[report]

Use it from any AI stack

narrativediff is interoperable with every popular way of using AI:

• MCP server — narrativediff mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)
• OpenAI-compatible / JSON — pipe narrativediff scan . --format json into any agent or LLM
• LangChain · CrewAI · AutoGen · LlamaIndex — wrap the CLI/JSON as a tool in one line
• CI / scripts — exit codes + SARIF for non-AI pipelines

How it compares

	Cognis narrativediff	Media-Bias-Group
Self-hostable, no account	✅	varies
Single command, zero config	✅	⚠️
JSON + SARIF for CI	✅	varies
MCP-native (AI agents)	✅	❌
Polyglot ports (JS/Go/Rust)	✅	❌
Open license	✅ COCL	varies

Built in the spirit of Media-Bias-Group/MBIB, re-framed the Cognis way. Missing a credit? Open a PR.

Integrations

Pipes into your stack: SARIF for code-scanning, JSON for anything, an MCP server (narrativediff mcp) for AI agents, and a webhook forwarder for SIEM/Slack/Jira. See docs/INTEGRATIONS.md.

Install — every way, every platform

pip install "git+https://github.com/cognis-digital/narrativediff.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/narrativediff.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/narrativediff.git" # uv
pip install cognis-narrativediff                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/narrativediff:latest --help        # Docker
brew install cognis-digital/tap/narrativediff                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/narrativediff/main/install.sh | sh

Linux	macOS	Windows	Docker	Cloud
scripts/setup-linux.sh	scripts/setup-macos.sh	scripts/setup-windows.ps1	docker run ghcr.io/cognis-digital/narrativediff	DEPLOY.md (AWS/Azure/GCP/k8s)

Related Cognis tools

• claimtrace — Misinformation provenance tracer — earliest-known appearance graph
• deepcheck — Lightweight synthetic-media detector with C2PA validation
• electionlens — Influence-operations pattern monitor for election periods

Explore the suite → 🗂️ all 170+ tools · ⭐ awesome-cognis · 🔗 cognis-sources · 🤖 uncensored-fleet · 🧠 engram

Contributing

PRs, new rules, and demo scenarios are welcome under the collaboration-pull model — see CONTRIBUTING.md and SECURITY.md.

⭐ If narrativediff saved you time, star it — it genuinely helps others find it.

Interoperability

{} composes with the 300+ tool Cognis suite — JSON in/out and a shared OpenAI-compatible /v1 backbone. See INTEROP.md for the suite map, composition patterns, and reference stacks.

License

Source-available under the Cognis Open Collaboration License (COCL) v1.0 — free for personal, internal-evaluation, research, and educational use; commercial / production use requires a license ([email protected]). See LICENSE.

---