Back to use cases
Recruiting & Onboarding

New-hire glossary for week one

The acronyms, project names, and internal jargon: defined from how people actually use them.

Best for
Onboarding pack
Primitives
OntologyVocabularyDefinitions
Token savings
44%

The prompt

You are building a glossary for a new hire joining the {{team_name}} team. Using Sentra: 1. Use the Ontology Agent's vocabulary inventory for this team — the acronyms, project names, internal product nicknames, and recurring jargon Sentra has indexed. 2. For each term, define it from how the team actually uses it: the canonical sentence pattern Sentra extracted, plus two real example Interactions (lightly redacted if needed) where the term appears. 3. Rank terms by usage frequency in the last 90 days so the new hire sees the most-common 30 first. 4. Flag any term that has multiple competing definitions and surface them all (a real signal worth fixing). Output: - A clean alphabetical glossary, top-30 first. - A "watch out" section listing terms with competing meanings.

See it work

Run it from

ClaudeCursorChatGPTAny agent · MCP

How it runs

  1. 01Copy the prompt into any agent connected to Sentra over MCP.
  2. 02The agent reads the company memory, scoped to what you are allowed to see.
  3. 03One grounded answer comes back, with every line traceable to its source.

Or automate it with Actions

The same prompt runs as a Sentra Action: on a schedule, or on a semantic trigger. Output lands in Slack or email, and every send waits for your approval.

Trigger idea: onboarding pack

Sentraover MCPDefined from how people actually use the words

Week-one glossary

Atlas

usage · 40+ mentions

The V3 replatform program, not a customer. "Atlas Friday" means its delivery checkpoint.

Gardener

usage

The nightly memory-reconciliation job. "Gardener caught it" means a stale fact was retired.

The Friday call

usage

The weekly customer-review meeting, unrelated to Atlas Friday. Yes, everyone confuses them in week one.

Northgate

A prospect, not the office building. Context disambiguates; the glossary saves you the guess.

The prompt above becomes this. Fictional data, real mechanics: every line cites the meeting, thread, or record it came from.

Without Sentra

  1. 01Pull the team's threads, docs, and meeting transcripts from the last 6 months.
  2. 02Read each to identify recurring acronyms, project nicknames, and internal jargon.
  3. 03Manually build a vocabulary list, deduplicating variants.
  4. 04For each term, read context to extract a canonical definition.
  5. 05Find two illustrative example interactions per term — skim and redact.
  6. 06Estimate usage frequency by counting occurrences.
  7. 07Flag any term with multiple competing definitions — surface conflicts.
  8. 08Compose the alphabetical glossary.

~90,000 tokens

With Sentra

  1. 01Pull Sentra's `Ontology` vocabulary inventory for the team — acronyms, project names, jargon already indexed.
  2. 02Each `Term` carries its canonical definition pattern and two illustrative `Interactions`.
  3. 03Rank by 90-day usage frequency.
  4. 04Flag terms with competing meanings — surfaced as conflicts.
  5. 05Compose the glossary.

~50,000 tokens

Book a DemoStart Free