Claude Code Agent Development
anthropics/claude-plugins-official
Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.
Write and run custom Pi prompt templates (slash commands) for this extension.
Install Prompt Template Authoring
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoring --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/prompt-template-authoring .claude/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.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/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .claude/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoringType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoring --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/prompt-template-authoring .agents/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .agents/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoring --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/prompt-template-authoring .cursor/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .cursor/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/nicobailon/pi-prompt-template-model.git --path skills/prompt-template-authoring--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoring --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/prompt-template-authoring .gemini/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .gemini/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoringInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/prompt-template-authoring .github/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .github/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install nicobailon/pi-prompt-template-model prompt-template-authoring --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nicobailon/pi-prompt-template-model.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/prompt-template-authoring .opencode/skills/prompt-template-authoring && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "prompt-template-authoring" agent skill from https://github.com/nicobailon/pi-prompt-template-model/tree/main/skills/prompt-template-authoring into .opencode/skills/prompt-template-authoring/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prompt-template-authoring", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
prompt-template-authoring- GitHub stars
- 321
- Token cost
- ~1.3k tokens
- SKILL.md length
- 432 words
- Files
- 1
- Skills in repo
- 1
- Repo updated
- First seen
- Licence
- MIT
At a glance
Write and run custom Pi prompt templates (slash commands) for this extension.
- Creating templates with model selection
- SKILL.md covers Where Templates Live, Minimal Template, Model Selection and Argument Substitution, plus 7 more sections
- Calls bash
- Deterministic pre-steps
What it does
Prompt Template Authoring is an agent skill from nicobailon/pi-prompt-template-model. Write and run custom Pi prompt templates (slash commands) for this extension. Use when creating templates with model selection, deterministic pre-steps, loops, chains, subagents, or best-of-N compare flows.
Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in AI & LLM Engineering, covering Prompt engineering, Subagents and Hooks and plugins. The repository describes itself as: Pi extension: Add model frontmatter to prompt templates for automatic model switching. The licence is MIT.
When your agent uses it
- Creating templates with model selection
- Deterministic pre-steps
- Best-of-N compare flows
Example prompts
- “/prompt-template-authoring”
What it can do on your machine
Read from SKILL.md and the folder at commit 6da2059. 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:
bash
From the folder's file list and the shell code blocks in SKILL.md.
Network
Links to these hosts (documentation or services it may open):
github.com
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
Prompt Template Authoring loads about 1.3k tokens when it runs. Until then it costs about 58 tokens; SKILL.md has 432 words of instructions outside code blocks.
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 passed
The automated check found no risky patterns in SKILL.md.
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 nicobailon/pi-prompt-template-model at commit 6da2059, republished under its MIT licence (© nicobailon). 432 words, ~1,319 tokens.
Download SKILL.mdSave it as
.claude/skills/prompt-template-authoring/SKILL.md (or your agent's skills folder).- name
- prompt-template-authoring
- description
- Write and run custom Pi prompt templates (slash commands) for this extension. Use when creating templates with model selection, deterministic pre-steps, loops, chains, subagents, or best-of-N compare flows.
Prompt Template Authoring
Use this skill when working on prompt templates for pi-prompt-template-model.
Templates are markdown files that register as slash commands.
Where Templates Live
~/.pi/agent/prompts/— user prompts (highest priority).pi/prompts/inside a trusted project — project-specific prompts
Extension examples/ are reference files only. Copy them to a prompt directory to register them.
Minimal Template
---
model: claude-sonnet-4-20250514
---
Your prompt body here.Save as my-command.md, restart Pi, run /my-command. Use description: for autocomplete text.
Model Selection
Omit model: to inherit the current session model. Otherwise:
model: claude-sonnet-4-20250514— specific modelmodel: claude-opus-4, gpt-5.4— fallback order (tries first, falls back to second if unavailable)model: claude-opus-4, gpt-5.4+rotate: true— cycle through list on each loop iteration
Argument Substitution
The prompt body can use placeholders:
$@— all arguments passed to the command$1,$2— specific positional arguments${@:1}— argument 1 and everything after
Deterministic Steps (Pre-LLM Execution)
Run a command or script before the LLM turn. The model only sees the output if you want it to.
Two equivalent forms. Don't mix them in the same prompt.
Shorthand form — top-level keys:
---
run: git status --short
handoff: always
---
Summarize the repo state.Nested form — under deterministic::
---
deterministic:
run: ./scripts/ship.sh
handoff: on-failure
timeout: 60000
---
Diagnose the failure and suggest a fix.Handoff controls when the LLM sees the result:
never— run, show result, done (no LLM turn)always— always hand result to modelon-failure— only hand off if command exits non-zeroon-success— only hand off if command exits zero
Execution forms:
run: command string— runs via/bin/bash -lcrun: {command: git, args: [status], shell: false}— explicit args, optional shellscript: ./script.shorscript: {path: ./script.sh, args: [--fast]}— run a file
Constraints:
- Only single prompt templates (no
chain,loop,subagent, orparallel) - Runtime flags
--loop,--subagent,--forkare rejected for deterministic prompts
Show full SKILL.md (169 more words)Show less
Subagent Delegation
Delegate to another Pi agent instead of running inline:
---
model: claude-sonnet-4-20250514
subagent: delegate # or true, or a specific agent name
inheritContext: true # fork conversation context (optional)
cwd: /absolute/path # working directory for the subagent (optional)
parallel: 3 # run 3 copies in parallel (optional)
---
$@Requires pi-subagents to be installed.
Loops
Run the prompt multiple times:
---
model: claude-sonnet-4-20250514
loop: 5 # run exactly 5 times
converge: true # stop early if no changes (default)
fresh: true # collapse context between iterations
---
$@Or at runtime: /command --loop 5, /command --loop (unlimited), or /command --loop=5 --fresh.
Chains
Chain templates declare a reusable pipeline:
---
chain: analyze -> fix -> test
chainContext: summary # pass step summaries to later delegated steps
---
$@Or use /chain-prompts analyze -> fix -> test at runtime. Chain templates ignore the body and model: field.
Model Conditionals
Show different content based on which model runs:
<if-model is="anthropic/*">
Use Claude-specific instructions.
<else>
Use default instructions.
</if-model>Supports exact IDs, provider/model-id pairs, wildcards (anthropic/*), and comma-separated combinations.
Best-of-N Compare
Run multiple workers, aggregate with reviewers, optionally apply final changes:
---
description: Best-of-N code review
bestOfN:
worktree: true # required if using finalApplier
workers:
- model: openai-codex/gpt-5.4-mini:low
count: 2
reviewers:
- model: anthropic/claude-sonnet-4-20250514:medium
finalApplier:
agent: delegate
model: anthropic/claude-sonnet-4-20250514:high
---
$@Runtime Flags
Override frontmatter at invocation:
--model=provider/model-id— use this model instead--subagent/--subagent=<name>/--subagent:<name>— force delegation--fork— force delegation with context fork--loop N/--loop=N/--loop— override loop count (unlimited if bare)--fresh— collapse context between iterations--no-converge— run all iterations even if no changes--cwd=/absolute/path— working directory override when the prompt supportscwd--chain-context— pass summaries to later delegated chain steps--worktree— use git worktrees for parallel delegated work
When stuck, check README.md and examples/best-of-n.md in this extension.
© nicobailon, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/prompt-template-authoring of nicobailon/pi-prompt-template-model.
Open the folder on GitHubat commit 6da2059
Compare with similar skills
Prompt Template Authoring 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Prompt Template Authoring this skillnicobailon/pi-prompt-template-model | 321 | — | ~1.3k | Automated safety check: Pass | MIT | |
| Claude Code Agent Developmentanthropics/claude-plugins-official | 38k | 7 repos | ~2.8k | Automated safety check: Pass | Apache-2.0 | |
| Pi AgentK-Dense-AI/scientific-agent-skills | 48k | 1 repos | ~2.1k | Automated safety check: Pass | MIT | |
| Wjs Evaling Voicedrop Promptsjianshuo/claude-skills | 131 | — | ~475 | Automated safety check: Pass | MIT | |
| Reviewing Agent Definitionsbitwarden/ai-plugins | 155 | — | ~2.4k | Automated safety check: Pass | Custom licence | |
| Subagent CreatorgreatSumini/cc-system | 438 | — | ~958 | Automated safety check: Pass | MIT |
Similar skills
- Official
Pi Agent
K-Dense-AI/scientific-agent-skills
Builds with and operates Pi, the minimal terminal coding harness.
Wjs Evaling Voicedrop Prompts
jianshuo/claude-skills
A skill your agent uses when 王建硕 wants to evaluate whether a change to VoiceDrop's 挖矿 system prompt is actually better than the live version — runs the local eval harness (golden fixtures ×…
- Official
Reviewing Agent Definitions
bitwarden/ai-plugins
Reviews Claude Code agent definition files for tool-access security, triggering quality, and system prompt clarity.
Subagent Creator
greatSumini/cc-system
Create specialized Claude Code sub-agents with custom system prompts and tool configurations.
Test Prompt
NeoLabHQ/context-engineering-kit
A skill your agent uses when creating or editing any prompt (commands, hooks, skills, subagent instructions) to verify it produces desired behavior - applies RED-GREEN-REFACTOR cycle to prompt…
Related
Categories
Questions about Prompt Template Authoring
What does Prompt Template Authoring do?
Write and run custom Pi prompt templates (slash commands) for this extension. Prompt Template Authoring is an agent skill from nicobailon/pi-prompt-template-model. Write and run custom Pi prompt templates (slash commands) for this extension.
When should I use Prompt Template Authoring?
Prompt Template Authoring fits situations like: creating templates with model selection; deterministic pre-steps; best-of-N compare flows.
How do I install Prompt Template Authoring in Claude Code?
Run `npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a claude-code`. Or copy the skill folder (skills/prompt-template-authoring in nicobailon/pi-prompt-template-model) into .claude/skills/prompt-template-authoring in your project. Claude Code loads it when a task matches its description.
How do I install Prompt Template Authoring in Codex?
Run `npx skills add nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a codex`. Or copy the skill folder (skills/prompt-template-authoring in nicobailon/pi-prompt-template-model) into .agents/skills/prompt-template-authoring in your project. Codex loads it when a task matches its description.
Can I use Prompt Template Authoring 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 nicobailon/pi-prompt-template-model --skill prompt-template-authoring -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/prompt-template-authoring, .gemini/skills/prompt-template-authoring, .github/skills/prompt-template-authoring and .opencode/skills/prompt-template-authoring in your project.
What does Prompt Template Authoring need to run?
Going by SKILL.md and its folder, Prompt Template Authoring needs the command-line tools its instructions call (bash).
Does Prompt Template Authoring access the network?
SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.
Is Prompt Template Authoring safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Prompt Template Authoring use?
Prompt Template Authoring 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 Prompt Template Authoring use?
About 1.3k tokens (SKILL.md is roughly 5.3k 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 Prompt Template Authoring?
Skills that share tags, products or a category with Prompt Template Authoring: Claude Code Agent Development (anthropics/claude-plugins-official, 38k stars), Pi Agent (K-Dense-AI/scientific-agent-skills, 48k stars), Wjs Evaling Voicedrop Prompts (jianshuo/claude-skills, 131 stars) and Reviewing Agent Definitions (bitwarden/ai-plugins, 155 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Prompt Template Authoring?
nicobailon (a GitHub user) maintains it in nicobailon/pi-prompt-template-model, which has 321 GitHub stars. The repository was last updated on September 21, 2026.
Source: nicobailon/pi-prompt-template-model on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.