The 100+ Native Actions
Reaching into Slack, Microsoft, Salesforce, Jira, GitHub, and Google
You'll be able to
- Pick the right native action for a step instead of a generic one
- Reason about the permissions a chosen action carries
Read first
Day 3 Lab
**Status: Generally available — Glean’s Agent Builder and its core features (scheduled triggers, versioning, native actions) are GA. Only the *content triggers* trigger type remains Beta; confirm availability in your tenant.**
- List the apps your chore touches: Slack, Jira, GitHub, Google, Salesforce, Microsoft.
- For each, find the native action that matches the step — a *read* before a *write*.
- For one write action, write down the permission it carries (who it acts as, what it can change).
- Replace any "do it manually" gap in your plan with the closest native action.
- Note any step where no native action fits — that's a flag, not a workaround.
Working Example: Choosing Native Actions Per Step
Per Glean's Fall '25 launch and developer-platform announcements, agents can call from a catalog of 100+ native actions across major apps. Verified against fall25_launch.
Goal: triage a new bug report and notify the owner.
Step 1 Jira -> get_issue (read) safe, read-only
Step 2 GitHub -> search_pull_requests (read) find the related PR
Step 3 Slack -> post_message (WRITE) posts as the agent's identity
Step 4 Salesforce-> update_case (WRITE) changes a customer record!
Pause at every WRITE: who does it act as, and what can it change?
Step 4 mutates customer data — that one earns a human check before it ships.The catalog is broad — Slack, Microsoft, Salesforce, Jira, GitHub, Google — so the work is *selection*, not plumbing. Reads are cheap to trust; writes carry real authority, so mapping each action to its permissions is where MAP lives. These actions are announced capability: confirm which ones are actually enabled and GA in your tenant before a plan relies on them.
Check for understanding
Before adding a write action, what single question about its permissions should you answer?
Check yourself
1. In Glean's vibe-code agent builder, how do you start building an agent?
2. The vibe-code builder, 100+ native actions, scheduled triggers, looping, and versioning are best described as...
3. Before adding a native WRITE action (e.g., update a Salesforce case) to an agent step, what should you check?
4. An agent runs unattended on a daily schedule. Which guard most directly keeps a bad run from compounding?
5. Why does rolling an agent back to a known-good version count as a governance control?
In 10 seconds
“Glean's Fall '25 launch announced 100+ native actions across Slack, Microsoft, Salesforce, Jira, GitHub, and Google. The skill is choosing the right one — and knowing what each is allowed to touch.”
Scheduled Triggers