See the stories forming in public, with the evidence.
Public Telegram channels and Reddit communities, read in their own language and grouped into the narratives that matter. Each one shows where it started, how fast it is moving, and whether an AI engine has picked it up.
Eidoscore monitors public Telegram channels and Reddit communities only, never private groups. A language model classifies each item for topic, stance and type, groups related items into narratives with severity, momentum and flags, and keeps the original text with a translation. Flags are early warnings for a human reviewer, never automated accusations.
Public Telegram and Reddit in one feed, read in their own language.
Every item is classified for stance, hostile, questioning, neutral or supportive, and linked to the narrative it belongs to. The original text stays, with a translation underneath.
- Views, shares, upvotes and replies kept with each item
- Filter by language, stance, channel or narrative
- Private groups are never collected
Topics and channels you set, with their health in view.
You define the topics, keywords, brands and pages to watch, and the public channels to collect from. Collection health is monitored, so a channel that goes private or a run that fails is flagged rather than mistaken for a quiet day.
- Telegram, Reddit and the AI engines, each on its own schedule
- Reliability and state on every source
- Failing sources sort to the top
Thousands of mentions, grouped into the stories that matter.
Each narrative carries a severity from Info to Critical, a momentum (rising, emerging, peaking, steady or fading) and a flag where a claim looks false, misleading, unverified or outdated. All of it is recalculated as new items arrive.
- Mentions, trend and change over seven days
- Channel and stance make-up at a glance
- A recommended next step and an owner on each narrative
Where it started, how it travelled, and when an engine repeated it.
A volume-by-channel view with the moments that changed a narrative's course marked on it: crossed to Reddit, first AI echo, raised to critical.
- Telegram, Reddit and AI engine mentions, day by day
- Estimated reach, languages and topic
- First seen to first AI echo, as a timeline
Every narrative links to the posts and answers behind it.
Open a narrative and read the key evidence as written, the sources ranked by mentions, reach and reliability, and what each AI engine answered and how often.
- Key evidence with the original text and a translation
- Sources ranked by share, reach, stance and reliability
- AI answers per engine, with the pages they cited
Flagged claims wait for a person. Nothing is accused automatically.
Items that look false, misleading or unverified go into a review queue ranked by risk, with the evidence and four scores attached. Approving a recommendation creates the task. Dismissing it is logged too.
- Risk, urgency, opportunity and confidence on every item
- Waiting items by type: misinformation, AI echo, unverified claim, outdated answer and more
- Flags are early warnings for a reviewer, never claims of coordination
See the narratives forming around your organisation.
A working session on your topics and channels, with real posts, answers and sources.