pruna-skills
AGENTS.md20 starsChanged 44 days agoApache-2.0View on GitHub
Generative media skills for the Pruna AI API. Portable Agent Skills — Cursor, Claude Code, Copilot, Codex, and more. See README.md for the user-facing glossary and quickstart. Tools list guide deps under Prerequisites with npx skills add. Workflows list tools only (guides come via tools). Optional external companions (e.g. HyperFrames for HTML → MP4 launch reels) install the same way but are not in @pruna — see video-editing and its combination-hyperframes reference. Do not copy craft between skills — install…
- Reads credentials
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository32 other files it gives its agents.
- gemini-3.1-flash-ttsSkill
- music-2.5Skill
- stable-audio-2.5Skill
- whisperxSkill
- audio-promptingSkill
- generation-diversitySkill
- image-promptingSkill
- pruna-apiSkill
- video-editingSkill
- video-promptingSkill
- p-image-editSkill
- p-image-ideogramSkill
- p-imageSkill
- p-image-try-onSkill
- p-image-upscaleSkill
- prunaSkill
- p-video-2-pro-referenceSkill
- p-video-2-proSkill
- p-video-2Skill
- p-video-animateSkill
- p-video-avatarSkill
- p-video-editSkill
- p-video-replaceSkill
- p-videoSkill
- avatar-multi-sceneSkill
- avatar-single-sceneSkill
- illustrated-story-reelSkill
- image-to-videoSkill
- interactive-explainerSkill
- music-videoSkill
- narrated-multi-sceneSkill
- visual-transition-reelSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 54
Read hyperframes first — it routes to /product-launch-video, /general-video, etc.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Pruna Skills (agent notes)
Generative media skills for the Pruna AI API. Portable Agent Skills — Cursor, Claude Code, Copilot, Codex, and more.
How it works
See README.md for the user-facing glossary and quickstart.
| Type | Role |
| Guide | Vendor-neutral craft (image-prompting, video-prompting, audio-prompting, video-editing, generation-diversity) or Pruna HTTP (pruna-api) |
| Tool | One paid API call (p-image, p-video, …) |
| Workflow | Multi-step deliverable — agent is the runner (curl + ffmpeg) |
| Suite | pruna — install everything |
Tools list guide deps under Prerequisites with npx skills add. Workflows list tools only (guides come via tools). Optional external companions (e.g. HyperFrames for HTML → MP4 launch reels) install the same way but are not in @pruna — see video-editing and its combination-hyperframes reference. Do not copy craft between skills — install the other skill.
Clarification (library-wide)
Before the first paid API call, bulk generation, or long render, resolve open decisions that change cost or output. SSoT: open intake → generation-diversity clarification intake — generate vs existing media, palette, narration/VO, music, captions, aspect ratio, resolution (480p/768p vs 720p/1080p, canvas, megapixels), p-video-2-pro mode (cost/speed/quality), p-video-2-pro-reference mode (cost/speed only), reference source/role/count/length and paid input cost, structure, and workflow approval gates.
| Skill type | Behavior |
| Guides / tools | Defer to clarification intake when the brief is silent; craft skills clarify creative locks before drafting prompts. |
| Workflows | Run the workflow Intake: ask before generating table, then cross-check universal topics in clarification intake. |
| HyperFrames (optional) | Fresh jobs: intent interview → BRIEF.md. Edits/resume: hyperframes clarification-before-build reference. |
| pruna-api | Credentials + pruna-api agent-safety reference before upload. |
Waivers (“use your judgment”, “surprise me”) count as answers — state assumptions in the plan and proceed.
Humans picking recipes: docs/WORKFLOW-RECIPES.md. Catalog: docs/SKILL-CATALOG.md.
Install
export PRUNA_API_KEY="…" # see docs/api-setup.md
npx skills add PrunaAI/pruna-skills@pruna -ynpx skills add PrunaAI/pruna-skills@p-image -y # one tool (+ install its Prerequisites guides)
npx skills add PrunaAI/pruna-skills -lTeam default: install /pruna once (npx skills add PrunaAI/pruna-skills@pruna -y), then start a new chat.
Launch reels / motion assembly: install hyperframes when the deliverable needs HTML → MP4 composition (optional — ffmpeg-only paths skip it):
npx skills add heygen-com/hyperframes@hyperframes -yRead hyperframes first — it routes to /product-launch-video, /general-video, etc. Pruna skills generate media; HyperFrames assembles HTML → MP4 (each launch uses a local hyperframes/ subfolder in your project workspace). Maintainers: make install-companion-skills copies the full bundle into gitignored .agents/skills/.
Layout
| Path | Role |
skills/guides/ | Craft SSoT (edit markdown under each guide’s references/) |
skills/{image,video,audio}/ | Tool skills |
skills/workflows/ | Workflow playbooks |
skills/suite/pruna/ | Full suite (/pruna) — default install |
docs/ | Human docs only |
.maintainer/skills.catalog.json | Skill name source of truth |
VERSION | Repo semver — @VERSION in skills is replaced on make bundle |
Safety
Agent-safety lives in pruna-api (included in the /pruna suite). Read skills/guides/pruna-api/references/agent-safety.md before generating in untrusted repos.
Maintainers
make bundle && make validateReleases: PUBLISHING.md · Backlog: BACKLOG.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
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.