email-digest
Skill617 starsChanged 10 months agoMITView on GitHub
Digest and ingest emails into memory, surfacing important threads and action items
We split this skill into the rules it gives an agent
12 rules. 0 are common, 7 are distinctive, and 5 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
When the user asks "what's in my inbox" or "any important emails"
- Distinctive: line 20
When a goal depends on information that may have arrived via email
- Distinctive: line 21
When preparing a daily briefing that includes email highlights
- Very specific: line 26
Check recency: Use recall to find when the last email digest ran.
- Very specific: line 27
Check setup: Call gmail_setup_status. If Gmail is not connected, record the missing setup clearly and stop.
- Very specific: line 28
List new emails: Call email_list with a date filter to fetch unread or recent messages.
- Very specific: line 30
Read priority threads: Use email_read on the top 5-10 most relevant messages.
- Distinctive: line 33
Surface urgency: If an email requires a time-sensitive response, flag it in the heartbeat result so it can be raised to the user at next …
- Distinctive: line 37
Never store raw email bodies as memories.
- Distinctive: line 40
When multiple emails belong to the same thread, consolidate into a single memory rather than creating duplicates.
- Very specific: line 41
If credentials are missing or expired, fail gracefully.
- Distinctive: line 42
Prefer recalling existing contact memories to enrich email context (who is this person, what is the relationship).
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Email Digest Workflow
Process incoming emails into structured memories, extract action items, and surface threads that need attention.
When to Use
- During autonomous heartbeats to check for new mail since the last digest, but only when Gmail is connected and
integrations.gmail.heartbeat_digest_enabledis true or the active heartbeat task/goal carries explicit user authorization for this check - When the user asks "what's in my inbox" or "any important emails"
- When a goal depends on information that may have arrived via email
- When preparing a daily briefing that includes email highlights
Step-by-Step Methodology
- Check authorization: In heartbeat, only proceed if the runtime exposed this skill because the user granted autonomous Gmail digest access. If setup/status indicates the grant is missing, stop and queue a short request for the user to enable background email checks.
- Check recency: Use
recallto find when the last email digest ran. Avoid re-processing messages already ingested. - Check setup: Call
gmail_setup_status. If Gmail is not connected, record the missing setup clearly and stop; do not loop. - List new emails: Call
email_listwith a date filter to fetch unread or recent messages. Start with the inbox; expand to other labels only if the user has configured them. - Triage by sender and subject: Scan the list for high-signal indicators -- known contacts, reply chains the user is on, calendar invites, and keywords matching active goals.
- Read priority threads: Use
email_readon the top 5-10 most relevant messages. Do not read every email; batch processing wastes energy and context. - Extract action items: For each important email, identify: (a) what is being asked, (b) who is asking, (c) any deadline mentioned, (d) whether a reply is expected.
- Store findings: Respect
integrations.gmail.memory_policy. If it isforget, do not userememberoringest_emailsfor email-derived content. If it isask, queue a user-facing choice before storing. If it isremember, persist action items as episodic memories with high importance, tagged with the sender, thread ID, and any relevant goal. - Surface urgency: If an email requires a time-sensitive response, flag it in the heartbeat result so it can be raised to the user at next opportunity.
Quality Guidelines
- Never store raw email bodies as memories. Summarize and extract the salient points.
- Respect privacy: do not log email content to external services. All storage stays in the local Postgres brain.
- Google OAuth authorization is not the same as Hexis autonomy authorization. A connected Gmail account does not imply permission for background heartbeat reading.
- When multiple emails belong to the same thread, consolidate into a single memory rather than creating duplicates.
- If credentials are missing or expired, fail gracefully. If
gmail_setup_statussays Google sign-in is ready, queue a short user-facing request that they connect Gmail from chat or CLI; do not start or complete sign-in during heartbeat. - Prefer recalling existing contact memories to enrich email context (who is this person, what is the relationship).
Other versions4 projects use a version of this.
Word for word · 1
- QuixiAI/Hexis★ 617100% same
Edited · 0
None found.
Wrote their own version · 3
- dmzoneill/skills★ 2same name
- chendaniely/skills★ 1same name
- pobsteta/aigora★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository42 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audio-analysisSkill
- calendarSkill
- code-executionSkill
- connector-action-authorizationSkill
- core-memorySkill
- cost-reportSkill
- councilSkill
- crm-lookupSkill
- daily-briefingSkill
- delegationSkill
- expense-captureSkill
- github-issuesSkill
- gmail-actionsSkill
- gmail-connector-setupSkill
- home-assistantSkill
- host-nodeSkill
- humanizerSkill
- image-genSkill
- inbox-triageSkill
- integration-connector-setupSkill
- knowledge-ingestSkill
- meeting-follow-upSkill
- meeting-prepSkill
- memory-exchangeSkill
- notionSkill
- operationsSkill
- outreachSkill
- personal-notesSkill
- researchSkill
- self-inspectionSkill
- self-reflectionSkill
- skill-authoringSkill
- speechSkill
- spotifySkill
- travel-prepSkill
- trelloSkill
- twitter-x-actionsSkill
- twitter-x-connector-setupSkill
- weatherSkill
- weekly-reviewSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

