ps-summary

name	ps:summary
description	Summarize any content — YouTube video, Medium article, Jira issue, GitHub repo, AWS/Amazon blog, or Slack thread. Auto-detects content type from URL or issue key.

ps:summary

Summarize any content — YouTube video, Medium article, Jira issue, GitHub repo, AWS/Amazon blog, or Slack thread. Auto-detects content type from URL or issue key.

Usage

/ps:summary <url-or-issue-key>

Examples:

• /ps:summary https://www.youtube.com/watch?v=abc123
• /ps:summary https://medium.com/some-article
• /ps:summary PROJ-123
• /ps:summary https://myorg.atlassian.net/browse/PROJ-123
• /ps:summary https://github.com/owner/repo
• /ps:summary https://aws.amazon.com/blogs/...
• /ps:summary https://myorg.slack.com/archives/C.../p...

---

Step 0 — Bootstrap

ls "$HOME/.local/share/personal-skills/scripts/summary/detect_type.py" 2>/dev/null && echo "ok" || echo "missing"

If missing:

curl -fsSL https://raw.githubusercontent.com/minhtranin/personal-skills/main/install.sh | bash

---

Step 1 — Detect content type

python3 "$HOME/.local/share/personal-skills/scripts/summary/detect_type.py" "<input>"

Output: youtube | medium | jira | github | amazon | slack | unknown

If unknown: tell the user "Could not detect content type for: <input>. Supported inputs: YouTube URL, Medium/Freedium URL, Jira issue key (e.g. PROJ-123) or URL, GitHub URL, AWS/Amazon blog URL, Slack thread URL." and stop.

---

Step 2 — Route to matching handler

Follow the section below that matches Step 1 output.

---

[youtube] YouTube Video

Y1 — Check deps:

bash "$HOME/.local/share/personal-skills/scripts/tube/check_deps.sh"

Y2 — Fetch title + transcript:

yt-dlp --get-title "<url>" 2>/dev/null || echo "Unknown Title"
bash "$HOME/.local/share/personal-skills/scripts/tube/get_transcript.sh" "<url>"

Y3 — Output summary:

Output a text summary in terminal:

YouTube Summary
└── <title>
    ├── URL     : <url>
    ├── Summary
    │   └── <3-5 sentences covering main topic and key takeaways>
    └── Key Points
        ├── • <point>
        └── ...

---

[medium] Medium Article

M1 — Fetch article:

python3 "$HOME/.local/share/personal-skills/scripts/medium/fetch_medium.py" "<url>"

M2 — Output summary:

Output a text summary in terminal:

Medium Summary
└── <title> — <author>
    ├── URL     : <url>
    ├── Summary
    │   └── <3-5 sentences covering main argument and conclusions>
    └── Key Points
        ├── • <point>
        └── ...

---

[jira] Jira Issue

J1 — Check credentials:

bash "$HOME/.local/share/personal-skills/scripts/jira/check_jira_credentials.sh"

Stop if missing.

J2 — Normalize — bare key (e.g. PROJ-123) used directly; URL → extract issue key from path.

J3 — Fetch issue:

python3 "$HOME/.local/share/personal-skills/scripts/jira/fetch_jira.py" "<issue-key>"

J3.1 — Interpret strikethrough text:

In the fetched description and comments, any text wrapped in ~~...~~ is strikethrough in Jira. Treat it as removed / superseded / no longer required — QC or the reporter struck it out instead of hard-deleting it. Rules:

• Do not list struck-through text as an active requirement, scope item, or key point.
• When struck text sits next to its replacement (a before→after edit), report only the current (non-struck) version, and note the change if it's material — e.g. "Requirement changed: was X, now Y."
• If a whole requirement/section is struck with no replacement, either omit it or mention it once as "Removed: …" so the reader knows it was dropped — never as something still to be done.

J3.5 — Inspect image attachments (if any):

If fetch_jira.py returned an attachments field with image files, download them locally:

python3 "$HOME/.local/share/personal-skills/scripts/jira/download_jira_images.py" '<attachments-json>'

Returns [{filename, mime_type, path}]. Inspect each downloaded image path with the current agent/model's native image understanding; do not call an external vision API or require GEMINI_API_KEY.

Use image findings only when they add useful context beyond the issue text, such as screenshots showing UI state, error messages, diagrams, affected screens, or expected behavior. Incorporate useful findings into Summary and Key Points, and add an Images section only when the image materially clarifies the requirement or bug. If images are decorative, duplicated, inaccessible, or not helpful, omit the section and continue.

J4 — Output summary as tree in terminal:

Jira Summary
└── [<TYPE>] <KEY>: <title>
    ├── Status   : <status>
    ├── Priority : <priority>
    ├── Assignee : <assignee>
    ├── Reporter : <reporter>
    ├── Summary
    │   └── <3-5 sentences>
    ├── Key Points
    │   ├── • <problem / scope / fix / blockers / decisions>
    │   └── ...
    ├── Images (optional, only if useful)
    │   ├── • <filename>: <useful visual context, if any>
    │   └── ...
    └── Comment Highlights
        ├── • <highlight>
        └── ...

---

[github] GitHub Repository

G1 — Fetch repo:

python3 "$HOME/.local/share/personal-skills/scripts/github/fetch_github_repo.py" "<url>"

G2 — Output summary:

Output a text summary in terminal:

GitHub Summary
└── <owner>/<repo>
    ├── URL         : <url>
    ├── Description : <description>
    ├── Tech Stack  : <languages / frameworks>
    ├── Summary
    │   └── <3-5 sentences covering purpose, architecture, notable patterns>
    └── Key Points
        ├── • <point>
        └── ...

---

[amazon] AWS / Amazon Blog

A1 — Fetch article:

python3 "$HOME/.local/share/personal-skills/scripts/amazon/fetch_amazon_blog.py" "<url>"

A2 — Output summary:

Output a text summary in terminal:

AWS Blog Summary
└── <title> — <author>
    ├── URL      : <url>
    ├── Category : <category>
    ├── Summary
    │   └── <3-5 sentences covering the AWS solution and key services used>
    └── Key Points
        ├── • <point>
        └── ...

---

[slack] Slack Thread

S1 — Check tokens:

bash "$HOME/.local/share/personal-skills/scripts/slack/check_slack_tokens.sh"

Stop if missing.

S2 — Fetch thread:

python3 "$HOME/.local/share/personal-skills/scripts/slack/fetch_slack_thread.py" "<url>"

Exit codes: 0 = success, 1 = auth expired (re-run check_slack_tokens.sh and retry), 2 = fatal error.

S3 — Output summary as tree in terminal:

Slack Summary
└── #<channel> thread
    ├── Replies      : <count>
    ├── Participants : <names>
    ├── Date         : <date>
    ├── Summary
    │   └── <3-5 sentences>
    ├── Key Points
    │   ├── • <decisions / action items / open questions>
    │   └── ...
    └── Participants
        ├── <name> — <role/contribution>
        └── ...