email-meeting-summary

Name: email-meeting-summary
Author: opendatahub-io

36stars74forks

How to install email-meeting-summary?

opendatahub-io/ai-helpers/email-meeting-summary

$npx -y skills add opendatahub-io/ai-helpers --skill email-meeting-summary

Installs into the current project.

Summary

Use when the user wants to summarize a Google Meet meeting and send the summary by email. Reviews a Google Meet transcript for a specific meeting topic, then composes a Gmail draft summarizing decisions and action items for that topic. Prompts for meeting selection if not specifi

Files · 1

View on GitHub

SKILL.md

1	# Email Meeting Summary
2
3	Read a Google Meet transcript, identify discussion topics, summarize decisions
4	and action items for the chosen topic, and compose a Gmail draft for review.
5
6	## Prerequisites
7
8	- `gws` binary must be on `$PATH`
9	- Authenticated: run `gws auth login` if not already logged in. The `--readonly` flag is sufficient for reading calendar/docs/drive, but draft creation requires the `gmail.compose` scope which is included in the default (non-readonly) login.
10
11	If `gws` is not installed or authentication fails, tell the user to follow the
12	setup instructions in the `google-workspace` skill.
13
14	Helper scripts are located in `${CLAUDE_SKILL_DIR}` and require Python 3.10+.
15
16	---
17
18	## Step 1: Identify the Meeting
19
20	Run the search once and save the result to a temp file:
21
22	```bash
23	EVENT_FILE=$(mktemp /tmp/meeting_event.XXXXXX)
24
25	# With a keyword (argument provided):
26	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 "<keyword>" > "$EVENT_FILE"
27
28	# With keyword + specific date (recurring meeting):
29	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 --date "<YYYY-MM-DD>" "<keyword>" > "$EVENT_FILE"
30
31	# No argument — list all recent events:
32	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 > "$EVENT_FILE"
33	```
34
35	The script always outputs JSON. Read `EVENT_FILE` and display the result(s) to
36	the user as a human-readable list.
37
38	The script searches the past 7 days (up to now). When a keyword matches
39	multiple occurrences of a recurring meeting, it automatically picks the best
40	one: the occurrence on `--date` if given, otherwise today's, otherwise the most
41	recent past one.
42
43	Always confirm with the user before proceeding, regardless of how many
44	results were found. Use `AskUserQuestion`:
45
46	> "I found: <title> on <date> at <time>. Is this the meeting you want to
47	> summarize? (yes / no — if no, describe the meeting you meant)"
48
49	If the user says no, ask for clarification and re-run `find_meeting.py` with a
50	corrected keyword or `--date`, overwriting `EVENT_FILE`. If multiple events were
51	returned, list them and ask the user to pick one by number, then re-run with
52	`--date` to narrow to that occurrence.
53
54	---
55
56	## Step 2: Find the Transcript in Google Drive
57
58	Pass the event JSON to `find_transcript.py` on stdin:
59
60	```bash
61	python3 "${CLAUDE_SKILL_DIR}/scripts/find_transcript.py" < "$EVENT_FILE"
62	```
63
64	The script returns JSON with a `source` field:
65
66	\| `source` \| Meaning \| What to use \|
67	\|---\|---\|---\|
68	\| `"attachment"` \| Found in event attachments \| `fileId` field \|
69	\| `"drive"` \| Found via Drive search \| pick from `files` array \|
70	\| `"none"` \| Not found anywhere \| inform user and stop \|
71
72	If `source` is `"none"`: tell the user the transcript is not yet available
73	(Gemini may take several minutes after the meeting ends) and stop.
74
75	If `source` is `"drive"` and `files` has more than one entry: present a
76	numbered list (name, created date, link) and use `AskUserQuestion` to ask which
77	file to use.
78
79	If exactly one result: proceed automatically.
80
81	Store the chosen document ID as `DOC_ID`.
82
83	---
84
85	## Step 3: Read the Transcript
86
87	```bash
88	python3 "${CLAUDE_SKILL_DIR}/scripts/read_doc.py" "<DOC_ID>"
89	```
90
91	The script outputs the document as plain text. Store this as `TRANSCRIPT`. If
92	the script exits with an error or the output is empty, inform the user and stop.
93
94	---
95
96	## Step 4: Identify Topics and Let the User Choose
97
98	Analyze `TRANSCRIPT` to identify 3–8 distinct discussion topics or agenda
99	items. For each topic, write a one-sentence description of what was discussed.
100
101	Present the topics as a numbered list, for example:
102
103	```text
104	1. Sprint planning — The team aligned on story point allocations for the
105	upcoming two-week sprint.
106	2. CI pipeline failures — Discussed root cause of the nightly test failures
107	and assigned investigation owners.
108	3. Q3 roadmap review — Reviewed progress against Q3 milestones and flagged
109	three items at risk.
110	```
111
112	Use `AskUserQuestion`:
113
114	> "Which topic would you like to summarize for the team email? Enter the number."
115
116	---
117
118	## Step 5: Generate the Summary and Present for Review
119
120	For the selected topic, extract from `TRANSCRIPT`:
121
122	Key decisions — concrete choices or conclusions reached (bullet list).
123	Action items — tasks assigned, including owner name and due date where
124	mentioned (bullet list; format each as `[Own

People who liked this also viewed

6 picks

Type

email-meeting-summary

36stars74forks

View on GitHub

How to install email-meeting-summary?

opendatahub-io/ai-helpers/email-meeting-summary

$npx -y skills add opendatahub-io/ai-helpers --skill email-meeting-summary

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Email Meeting Summary
2
3	Read a Google Meet transcript, identify discussion topics, summarize decisions
4	and action items for the chosen topic, and compose a Gmail draft for review.
5
6	## Prerequisites
7
8	- `gws` binary must be on `$PATH`
9	- Authenticated: run `gws auth login` if not already logged in. The `--readonly` flag is sufficient for reading calendar/docs/drive, but draft creation requires the `gmail.compose` scope which is included in the default (non-readonly) login.
10
11	If `gws` is not installed or authentication fails, tell the user to follow the
12	setup instructions in the `google-workspace` skill.
13
14	Helper scripts are located in `${CLAUDE_SKILL_DIR}` and require Python 3.10+.
15
16	---
17
18	## Step 1: Identify the Meeting
19
20	Run the search once and save the result to a temp file:
21
22	```bash
23	EVENT_FILE=$(mktemp /tmp/meeting_event.XXXXXX)
24
25	# With a keyword (argument provided):
26	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 "<keyword>" > "$EVENT_FILE"
27
28	# With keyword + specific date (recurring meeting):
29	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 --date "<YYYY-MM-DD>" "<keyword>" > "$EVENT_FILE"
30
31	# No argument — list all recent events:
32	python3 "${CLAUDE_SKILL_DIR}/scripts/find_meeting.py" --days 7 > "$EVENT_FILE"
33	```
34
35	The script always outputs JSON. Read `EVENT_FILE` and display the result(s) to
36	the user as a human-readable list.
37
38	The script searches the past 7 days (up to now). When a keyword matches
39	multiple occurrences of a recurring meeting, it automatically picks the best
40	one: the occurrence on `--date` if given, otherwise today's, otherwise the most
41	recent past one.
42
43	Always confirm with the user before proceeding, regardless of how many
44	results were found. Use `AskUserQuestion`:
45
46	> "I found: <title> on <date> at <time>. Is this the meeting you want to
47	> summarize? (yes / no — if no, describe the meeting you meant)"
48
49	If the user says no, ask for clarification and re-run `find_meeting.py` with a
50	corrected keyword or `--date`, overwriting `EVENT_FILE`. If multiple events were
51	returned, list them and ask the user to pick one by number, then re-run with
52	`--date` to narrow to that occurrence.
53
54	---
55
56	## Step 2: Find the Transcript in Google Drive
57
58	Pass the event JSON to `find_transcript.py` on stdin:
59
60	```bash
61	python3 "${CLAUDE_SKILL_DIR}/scripts/find_transcript.py" < "$EVENT_FILE"
62	```
63
64	The script returns JSON with a `source` field:
65
66	\| `source` \| Meaning \| What to use \|
67	\|---\|---\|---\|
68	\| `"attachment"` \| Found in event attachments \| `fileId` field \|
69	\| `"drive"` \| Found via Drive search \| pick from `files` array \|
70	\| `"none"` \| Not found anywhere \| inform user and stop \|
71
72	If `source` is `"none"`: tell the user the transcript is not yet available
73	(Gemini may take several minutes after the meeting ends) and stop.
74
75	If `source` is `"drive"` and `files` has more than one entry: present a
76	numbered list (name, created date, link) and use `AskUserQuestion` to ask which
77	file to use.
78
79	If exactly one result: proceed automatically.
80
81	Store the chosen document ID as `DOC_ID`.
82
83	---
84
85	## Step 3: Read the Transcript
86
87	```bash
88	python3 "${CLAUDE_SKILL_DIR}/scripts/read_doc.py" "<DOC_ID>"
89	```
90
91	The script outputs the document as plain text. Store this as `TRANSCRIPT`. If
92	the script exits with an error or the output is empty, inform the user and stop.
93
94	---
95
96	## Step 4: Identify Topics and Let the User Choose
97
98	Analyze `TRANSCRIPT` to identify 3–8 distinct discussion topics or agenda
99	items. For each topic, write a one-sentence description of what was discussed.
100
101	Present the topics as a numbered list, for example:
102
103	```text
104	1. Sprint planning — The team aligned on story point allocations for the
105	upcoming two-week sprint.
106	2. CI pipeline failures — Discussed root cause of the nightly test failures
107	and assigned investigation owners.
108	3. Q3 roadmap review — Reviewed progress against Q3 milestones and flagged
109	three items at risk.
110	```
111
112	Use `AskUserQuestion`:
113
114	> "Which topic would you like to summarize for the team email? Enter the number."
115
116	---
117
118	## Step 5: Generate the Summary and Present for Review
119
120	For the selected topic, extract from `TRANSCRIPT`:
121
122	Key decisions — concrete choices or conclusions reached (bullet list).
123	Action items — tasks assigned, including owner name and due date where
124	mentioned (bullet list; format each as `[Own

People who liked this also viewed

6 picks

Type