gemini-media
Base skill: understand any local audio or video file by sending it to Gemini Flash-Lite (native multimodal — transcribes speech, reads on-screen text/slides…
Skill by vitoliu93
Chalk Labs
Need help deploying gemini-media?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code. Read the repository before you give any tool access to your data.
npx agentmarket install skill/vitoliu93-gemini-media
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: presentations
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/gemini-media
Tools it pre-approves 5
From allowed-tools in SKILL.md: the agent may use these without asking you each time.
Bash(python3:*),Bash(yt-dlp:*),Bash(bili:*),Bash(ffmpeg:*),Task
What's inside
- Skill name
gemini-media- Description
- Base skill: understand any local audio or video file by sending it to Gemini Flash-Lite (native multimodal — transcribes speech, reads on-screen text/slides, summarizes) cheaply and dependency-free. Trigger on "理解这个视频/音频文件", "转写这个视频/录音", "用 gemini 看/听这个文件", "transcribe this video/audio", "summarize this media file", "what is said in this recording", or whenever a media file needs to become text. Also the designated **no-subtitle fallback** for the youtube-cli and bilibili-cli skills. Heavy media reading happens inside Gemini, so the raw audio/transcript never fills the main context.
- License
- unknown
- Allowed tools
- Bash(python3:*),, Bash(yt-dlp:*),, Bash(bili:*),, Bash(ffmpeg:*),, Task
- Instructions
- 77 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
- Pre-approves some shell commands
- Instructions call the network
Files (3)
- SKILL.md
- workflow.md · references/
- gemini_media.py · scripts/
The license isn't open source, so the instructions aren't shown here. Read SKILL.md on GitHub
More from vitoliu93/harness-dev-plugins
- 0.565score
ceo-mode
vitoliu93
Hold the CEO seat: keep the goal and the decisions, hand every hands-on step to agents, and report the outcome.
SkillAFreshness: AClaude Code +9 - 0.565score
ccobs
vitoliu93
Build or query the agent observability ledger obs.db, and roll its distilled rules into per-project digests.
SkillAFreshness: AClaude Code +9 - 0.565score
orchestrate
vitoliu93
Pick the scene team from the user's teams directory, compose its roles, and coordinate their work through completion.
SkillAFreshness: AClaude Code +9 - 0.524score
advanced-plan
vitoliu93
Write and track a deterministic, acceptance-bearing dev plan as a mini-project.
SkillBFreshness: BClaude Code +9 - 0.524score
debrief
vitoliu93
Archive plan artifacts, distill this session into ccobs, correct a rule that proved wrong, then improve, merge, or retire existing skills before registering…
SkillBFreshness: BClaude Code +9 - 0.524score
rate-limit-watcher
vitoliu93
Watch running Herdr agents for a subscription rate-limit wall and either wake them after reset or hand the task to another agent.
SkillBFreshness: BClaude Code +9 - 0.521score
grill-me
vitoliu93
Interview the user round by round across a design tree until every branch is settled.
SkillBFreshness: BClaude Code +9 - 0.519score
cto-audit
vitoliu93
Audit project architecture, domain model, and harness rules from a CTO lens.
SkillBFreshness: BClaude Code +9 - 0.519score
context-audit
vitoliu93
Audit always-loaded context or project docs, then adopt placement rules.
SkillBFreshness: BClaude Code +9 - 0.474score
youtube-cli
vitoliu93
This skill should be used to search YouTube and extract what videos actually say — via yt-dlp search plus subtitle/transcript extraction — then judge content…
SkillCFreshness: CClaude Code - 0.474score
bilibili-cli
vitoliu93
This skill should be used to search Bilibili (B站/哔哩哔哩) and extract what videos actually say — via the `bili` CLI's search plus one-shot subtitle + AI summary…
SkillCFreshness: CClaude Code - 0.474score
tech-selection
vitoliu93
Treat an open question — "which approach/library should we use?", "is this feasible?", "how do others solve this?" — as a tracked study that survives context…
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jun 26, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- License unknown
- Repository
- github.com/vitoliu93/harness-dev-plugins
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from vitoliu93
- 0.565score
orchestrate
vitoliu93
Pick the scene team from the user's teams directory, compose its roles, and coordinate their work through completion.
SkillAFreshness: AClaude Code +9 - 0.565score
ccobs
vitoliu93
Build or query the agent observability ledger obs.db, and roll its distilled rules into per-project digests.
SkillAFreshness: AClaude Code +9 - 0.565score
ceo-mode
vitoliu93
Hold the CEO seat: keep the goal and the decisions, hand every hands-on step to agents, and report the outcome.
SkillAFreshness: AClaude Code +9 - 0.524score
debrief
vitoliu93
Archive plan artifacts, distill this session into ccobs, correct a rule that proved wrong, then improve, merge, or retire existing skills before registering…
SkillBFreshness: BClaude Code +9 - 0.524score
advanced-plan
vitoliu93
Write and track a deterministic, acceptance-bearing dev plan as a mini-project.
SkillBFreshness: BClaude Code +9 - 0.524score
rate-limit-watcher
vitoliu93
Watch running Herdr agents for a subscription rate-limit wall and either wake them after reset or hand the task to another agent.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get gemini-media running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
gemini-media skill. Agent Mart. https://agentmart.dev/skills/vitoliu93-gemini-media (last updated 2026-10-24)