Agent skill

Perplexity Server Deploy

by escapeWu in escapeWu/perplexity-ai

Deploys the perplexity-ai project to its production server by pushing main, fast-forwarding the server checkout, rebuilding the image there and verifying health.

MITAuto-check: notesDevOps & Cloud

Install Perplexity Server Deploy

skills CLI
$ npx skills add escapeWu/perplexity-ai --skill server-deploy -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install escapeWu/perplexity-ai server-deploy --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/escapeWu/perplexity-ai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/server-deploy .claude/skills/server-deploy && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
server-deploy
GitHub stars
169
Token cost
~662 tokens
SKILL.md length
234 words
Files
2
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

Deploys the perplexity-ai project to its production server by pushing main, fast-forwarding the server checkout, rebuilding the image there and verifying health.

  • Works in 5 steps: Confirm the local branch is main, review… → Record the local commit and confirm… → Require a clean remote tracked worktree,… → …
  • Rolling out the latest main of perplexity-ai to the production server
  • SKILL.md covers Target, Workflow and Invariants
  • Calls git, ssh and docker

What it does

The skill fixes a single deployment path for one project and tells the agent to follow it exactly. The agent confirms the local branch is `main`, reviews the diff, commits only the requested files and pushes, then checks that `origin/main` matches the local commit. Over SSH it requires the server's tracked worktree to be clean, fast-forwards the checkout and confirms the remote commit equals the local one. It then runs `./deploy/compose.sh up` on the server, which builds the image from that source, replaces the running container and waits for health, and finally runs a verify step again.

The invariants are strict. The image is always built from the server checkout and never pulled from Docker Hub, the server `.env`, `token_pool_config.json`, `data/` and Docker volumes are preserved, and a dirty remote worktree stops the run. The agent never force-pushes, resets, cleans, runs `docker compose down` or deletes images and volumes, stops on any failed push, mismatch, build or check, and adds no attribution trailers unless you ask. It reports the deployed commit, image ID, `/health` result and final container state.

When your agent uses it

  • Rolling out the latest main of perplexity-ai to the production server
  • Rebuilding and restarting the service container after a code change
  • Confirming which commit is live and whether the service is healthy

Example prompts

  • “Deploy the latest main of perplexity-ai to the production server.”
  • “Rebuild and restart the server container, then report the deployed commit.”
  • “Roll out my last push and tell me the /health result.”

Requirements

  • SSH access to the configured production server
  • Push access to the project's origin on GitHub
  • Docker on the server with the deploy/compose.sh script

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Confirm the local branch is main, review the intended diff, commit only the
  2. Record the local commit and confirm origin/main resolves to the same SHA.
  3. Require a clean remote tracked worktree, fast-forward the server checkout,
  4. Build the application image from that checked-out source on the server,
  5. Recheck health and status after deployment.

What it can do on your machine

Read from SKILL.md and the folder at commit a39a13f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git
    • ssh
    • docker

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, ssh and docker, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Perplexity Server Deploy loads about 662 tokens when it runs. Until then it costs about 54 tokens; SKILL.md has 234 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~54
When it runs · the whole SKILL.md, loaded when a task matches
~662

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:71
    - Preserve the server `.env`, `token_pool_config.json`, `data/`, and Docker volumes.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from escapeWu/perplexity-ai at commit a39a13f, republished under its MIT licence (© escapeWu). 234 words, ~662 tokens.

Download SKILL.mdSave it as .claude/skills/server-deploy/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
server-deploy
description
Deploy perplexity-ai to its configured production server from the latest GitHub main source. Use when the user asks to deploy, publish, update, rebuild, restart, or roll out this project on the server.

Server Deploy

Target

  • Local Git remote: origin
  • Branch: main
  • SSH host: oralce-chunchuan
  • Server directory: /root/perplexity
  • Remote Git remote: origin
  • Remote deployment entrypoint: ./deploy/compose.sh

Workflow

Perform the following sequence exactly.

  1. Confirm the local branch is main, review the intended diff, commit only the requested files, and push the exact commit.

    bash
    git push origin main
  2. Record the local commit and confirm origin/main resolves to the same SHA. Stop if the push or equality check fails.

    bash
    git rev-parse HEAD
    git ls-remote origin refs/heads/main
  3. Require a clean remote tracked worktree, fast-forward the server checkout, and verify that the remote commit equals the local commit. Substitute the recorded full SHA for <commit>.

    bash
    ssh oralce-chunchuan \
      "cd /root/perplexity && \
       test -z \"\$(git status --porcelain --untracked-files=no)\" && \
       git pull --ff-only origin main && \
       test \"\$(git rev-parse HEAD)\" = '<commit>'"
  4. Build the application image from that checked-out source on the server, replace the running container, wait for health, and verify the service.

    bash
    ssh oralce-chunchuan \
      'cd /root/perplexity && ./deploy/compose.sh up'
  5. Recheck health and status after deployment.

    bash
    ssh oralce-chunchuan \
      'cd /root/perplexity && \
       ./deploy/compose.sh verify && \
       ./deploy/compose.sh status && \
       docker inspect perplexity-mcp \
         --format "image={{.Image}} status={{.State.Status}} health={{.State.Health.Status}}"'

Invariants

  • Build the application from the server checkout. Do not pull, publish, or wait for a Docker Hub application image.
  • Preserve the server .env, token_pool_config.json, data/, and Docker volumes.
  • Stop when the remote tracked worktree is dirty; inspect and preserve changes instead of overwriting them.
  • Never force-push, run git reset or git clean, run docker compose down, or delete images or volumes.
  • Stop on a failed push, commit mismatch, build, health check, or status check.
  • Do not add authorship or attribution trailers unless the user explicitly requests them and provides the exact identity.

Report the deployed commit, image ID, /health result, and final container state.

© escapeWu, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/server-deploy of escapeWu/perplexity-ai.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit a39a13f

Compare with similar skills

Perplexity Server Deploy next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Perplexity Server Deploy compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Perplexity Server Deploy this skillescapeWu/perplexity-ai169—~662Automated safety check: NotesMIT
Reflexo ReleaseMyriad-Dreamin/typst.ts1.2k—~1.5kAutomated safety check: PassApache-2.0
1panel App Builderarch3rPro/1Panel-Appstore213—~1.3kAutomated safety check: PassMIT
Releasebmeares/Meerschaum154—~1.1kAutomated safety check: NotesApache-2.0
Demo Local Rolloutcarverauto/serviceradar921—~4.2kAutomated safety check: PassApache-2.0
Buzz Self Hostingtonbistudio/buzz-skills274—~2.2kAutomated safety check: NotesMIT

Similar skills

  • Reflexo Release

    Myriad-Dreamin/typst.ts

    Guide Reflexo/typst.ts release preparation and operator handoffs.

    1.2k GitHub stars~1.5k tokensUpdated 12 days ago
    DevOps & CloudAuto-check passed
  • 1panel App Builder

    arch3rPro/1Panel-Appstore

    A skill your agent uses when packaging Docker deployments as 1Panel local app store apps, including GitHub projects, docker-compose.yml files, docker run commands, app metadata, version directories…

    213 GitHub stars~1.3k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Release

    bmeares/Meerschaum

    Meerschaum release process — bump version, update changelog, stage dev→main PR, run CI, publish to PyPI, tag, GitHub release, build/push Docker images, rebuild docs on prod VPS.

    154 GitHub stars~1.1k tokensUpdated 28 days ago
    DevOps & CloudAuto-check: notes
  • Demo Local Rollout

    carverauto/serviceradar

    Build unpublished sha-... An agent skill from carverauto/serviceradar.

    921 GitHub stars~4.2k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Buzz Self Hosting

    tonbistudio/buzz-skills

    A skill your agent uses when helping a user set up, debug, or operate a self-hosted Buzz relay through Docker Compose.

    274 GitHub stars~2.2k tokensUpdated 2 mo ago
    DevOps & CloudAuto-check: notes
  • Release

    ar-io/ar-io-node

    Drive the AR.IO Node release process end-to-end — preflight checks, prepare commit, finalize with image SHAs, test docker compose profiles, tag & publish, and post-release cleanup.

    127 GitHub stars~4.2k tokensUpdated today
    DevOps & CloudAuto-check: notes

More from escapeWu/perplexity-ai

  • Perplexity Search

    escapeWu/perplexity-ai

    Searches the live web with citations through perplexity-mcp v2 tools or a bundled Python REST client, with focused ask, deep research and detached tasks.

    169 GitHub stars~1.8k tokensUpdated 10 days ago
    Auto-check passed
  • Version Bump and Release

    escapeWu/perplexity-ai

    Bumps a project version, updates both changelogs, builds the frontend, then commits, tags and pushes the release.

    169 GitHub stars~528 tokensUpdated 10 days ago
    Auto-check passed

Works with

Categories

Questions about Perplexity Server Deploy

What does Perplexity Server Deploy do?

Deploys the perplexity-ai project to its production server by pushing main, fast-forwarding the server checkout, rebuilding the image there and verifying health. The skill fixes a single deployment path for one project and tells the agent to follow it exactly. The agent confirms the local branch is `main`, reviews the diff, commits only the requested files and pushes, then checks that `origin/main` matches the local commit.

When should I use Perplexity Server Deploy?

Perplexity Server Deploy fits situations like: rolling out the latest main of perplexity-ai to the production server; rebuilding and restarting the service container after a code change; confirming which commit is live and whether the service is healthy.

How do I install Perplexity Server Deploy in Claude Code?

Run `npx skills add escapeWu/perplexity-ai --skill server-deploy -a claude-code`. Or copy the skill folder (.agents/skills/server-deploy in escapeWu/perplexity-ai) into .claude/skills/server-deploy in your project. Claude Code loads it when a task matches its description.

How do I install Perplexity Server Deploy in Codex?

Run `npx skills add escapeWu/perplexity-ai --skill server-deploy -a codex`. Or copy the skill folder (.agents/skills/server-deploy in escapeWu/perplexity-ai) into .agents/skills/server-deploy in your project. Codex loads it when a task matches its description.

Can I use Perplexity Server Deploy in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add escapeWu/perplexity-ai --skill server-deploy -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/server-deploy, .gemini/skills/server-deploy, .github/skills/server-deploy and .opencode/skills/server-deploy in your project.

What does Perplexity Server Deploy need to run?

Going by SKILL.md and its folder, Perplexity Server Deploy needs the command-line tools its instructions call (git, ssh and docker). Our summary lists: SSH access to the configured production server; Push access to the project's origin on GitHub; Docker on the server with the deploy/compose.sh script.

Does Perplexity Server Deploy access the network?

SKILL.md contains no URLs. Its commands use git, ssh and docker, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Perplexity Server Deploy safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Perplexity Server Deploy use?

Perplexity Server Deploy is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Perplexity Server Deploy use?

About 662 tokens (SKILL.md is roughly 2.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Perplexity Server Deploy?

Skills that share tags, products or a category with Perplexity Server Deploy: Reflexo Release (Myriad-Dreamin/typst.ts, 1.2k stars), 1panel App Builder (arch3rPro/1Panel-Appstore, 213 stars), Release (bmeares/Meerschaum, 154 stars) and Demo Local Rollout (carverauto/serviceradar, 921 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Perplexity Server Deploy?

escapeWu (a GitHub user) maintains it in escapeWu/perplexity-ai, which has 169 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on September 27, 2026.

Source: escapeWu/perplexity-ai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.