Install
$ agentstack add skill-gokulb20-crewm8-social-media-manager-skill-graph-podcast-to-clips ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Podcast to Clips
Distill podcast and video interview transcripts into platform-native social assets. A single 60-minute episode should yield 5-8 posts across platforms, not just one "new episode out" announcement.
Purpose
Podcasts are a content goldmine that most brands under-leverage. Each episode contains 5-8 moments worth posting. This skill extracts them, formats them per platform, and sequences the clips so one episode feeds a week of content.
When to Use
- New podcast episode published
- Founder was a guest on someone else's podcast
- Older episode that performed well deserves a second push
- Building a clips library for ongoing content
Inputs Required
- Full podcast transcript (auto-generated is fine)
- Episode title, host/guest handles, episode URL
- Brand voice guide
- Any specific clips the founder wants to highlight
Quick Reference
| Podcast Segment | Best Clip Format | Best Platform | |----------------|-----------------|--------------| | Hot take / contrarian opinion | Quote tweet or text-over-video | X, TikTok | | Data point or surprising stat | Quote card + stat-led post | X, LinkedIn | | Personal story | Video clip (30-60s) + caption | IG, TikTok | | Framework or system | Carousel or thread | LinkedIn, X | | Practical tip / how-to | Short video (15-30s) or text post | TikTok, IG, X |
Procedure
- Read the full transcript. Timestamp every moment that:
- Made you pause or think differently
- Contained a specific number or data point
- Told a concrete story
- Stated something contrarian or controversial
- Offered a framework or step-by-step process
- Was genuinely funny or human
- For each timestamped moment, determine the best format:
- Quote tweet: exact quote + 1-2 lines of context
- Video clip: exact time range (15-90 seconds), text overlay hook
- Carousel/post: if it's a framework, draft slide structure
- Thread: if connected insights, build 3-5 tweet mini-thread
- For video clips:
- Clip length: 15-30s for TikTok/Reels, 30-90s for X, 60s max for LinkedIn
- Include captions/subtitles
- Write text overlay hook for first 1.5 seconds
- Write platform-native caption
- For each clip, draft: hook, caption/body text, CTA, hashtags.
- Organize clips into a posting schedule:
- Best clip first (within 24h of episode drop)
- Spread remaining clips across 5-7 days
- Do not post all clips the same day
- Include attribution. Tag the host's account (if guesting). Link to full episode in every clip's caption or final tweet.
Output Format
# Podcast Clips: [Episode Title]
## Source
[Episode link + timestamp]
## Clip Inventory
### Clip 1: [Topic]
**Timestamp:** [MM:SS - MM:SS]
**Format:** [Video clip / Quote tweet / Thread / Carousel]
**Platform:** [Primary + secondary]
**Hook / Text overlay:** [Text]
**Caption / Body:** [Full text]
**CTA:** [CTA]
**Hashtags:** [Platform-appropriate set]
### Clip 2...
...
## Posting Schedule
- Day 1: [Clip X on platform Y]
- Day 2: [Clip X on platform Y]
...
Done Criteria
The skill is complete when:
- At least 5 clips are extracted per 60-minute episode
- Each clip has a timestamp, format, platform assignment, and caption
- Video clips specify time ranges and text overlay hooks
- A posting schedule spaces clips across at least 5 days
- All clips include attribution links
Pitfalls
- Clips the host finds interesting but the audience doesn't
- Video clips without captions/subtitles (unwatchable for most consumption)
- Posting all clips at once (floods feed, reduces per-clip engagement)
- Same caption for every clip ("new episode out!") instead of clip-specific hooks
- Clips that require full-episode context to make sense
Verification
For each clip: if you saw this in your feed without knowing the podcast, would it make sense? If the hook is "new episode" — rewrite to a content-first hook.
Example
Input: 60-minute podcast episode. Founder guests on "The B2B Growth Show." Topic: cold outreach strategies.
Extracted clips:
- (Quote tweet) Timestamp 12:30 — "Most cold outreach is lazy, it's not broken." Format: quote tweet. Platform: X.
- (Framework) Timestamp 24:00 — "The 3-question qualification framework." Format: carousel. Platform: LinkedIn.
- (Story) Timestamp 38:00 — "How one cold email led to a $50K deal." Format: video clip (45s). Platform: IG + TikTok.
- (Data point) Timestamp 51:00 — "Open rates dropped 40% when we made this one change." Format: stat-led post. Platform: X.
- (Tip) Timestamp 55:00 — "The exact DM template that books meetings." Format: short video (20s). Platform: TikTok.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: gokulb20
- Source: gokulb20/Crewm8-Social-Media-Manager-Skill-Graph
- License: MIT
- Homepage: https://crewm8.ai
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.