New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Agent Authoring

by a-streetcoder in a-streetcoder/agent-deck

A skill your agent uses when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment.

MITAuto-check passedAgent Workflows

Install Agent Authoring

skills CLI
$ npx skills add a-streetcoder/agent-deck --skill agent-authoring -a claude-code

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

GitHub CLI
$ gh skill install a-streetcoder/agent-deck agent-authoring --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/a-streetcoder/agent-deck.git skills-src && mkdir -p .claude/skills && cp -r skills-src/agent-deck/bundled-skills/agent-authoring .claude/skills/agent-authoring && 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
agent-authoring
GitHub stars
118
Token cost
~1.4k tokens
SKILL.md length
672 words
Files
1
Skills in repo
5
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment.

  • Works in 9 steps: Scope: user-global, library/catalog,… → Role: explorer, planner, coder,… → Routing: write whenToUse as one concise… → …
  • Reviewing Agent Deck agents
  • SKILL.md covers Agent file shape, Prompt composition, Required decisions and Scope rules, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Agent Authoring is an agent skill from a-streetcoder/agent-deck. Use when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment.

Its SKILL.md is about 1.4k 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 Agent Workflows. The licence is MIT.

When your agent uses it

  • Reviewing Agent Deck agents
  • Including frontmatter
  • Supervisor behavior
  • Continuation behavior

Example prompts

  • “/agent-authoring”

Workflow steps

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

  1. Scope: user-global, library/catalog, builtin override, or builtin replacement. Ask the user before writing files when the intended scope…
  2. Role: explorer, planner, coder, reviewer, tester, docs writer, release helper, etc.
  3. Routing: write whenToUse as one concise sentence that tells the parent exactly when to delegate to this agent; keep it distinct from the…
  4. Tool boundary: prefer read, grep, find, ls; add bash, edit, write only when needed.
  5. Default outcome: set defaultExpectedOutcome to reportOnly, editFilesInWorktree, writeProjectFile, or directProjectWrites. Use…
  6. Supervisor behavior: include contact_supervisor only when the child should ask for decisions or meaningful blockers.
  7. Continuation behavior: native subagents start fresh by default; the parent can explicitly continue a prior Subagent ID for direct…
  8. Skills: assign explicit skill names in skills:. Agent Deck passes them through Pi native --skill injection. Do not use inheritSkills.
  9. Validation: specify what files, commands, or evidence the agent should inspect before completion.

What it can do on your machine

Read from SKILL.md and the folder at commit 9efbe6c. 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

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

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

  • Network

    No URLs in SKILL.md.

    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

Agent Authoring loads about 1.4k tokens when it runs. Until then it costs about 40 tokens; SKILL.md has 672 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~40
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

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 a-streetcoder/agent-deck at commit 9efbe6c, republished under its MIT licence (© a-streetcoder). 672 words, ~1,435 tokens.

Download SKILL.mdSave it as .claude/skills/agent-authoring/SKILL.md (or your agent's skills folder).
name
agent-authoring
description
Use when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment.

Agent Deck Agent Authoring

Use this skill when creating or reviewing Agent Deck agent markdown files.

Agent file shape

Agents are Markdown files with YAML frontmatter and a compact role prompt body:

markdown
---
name: example-agent
description: Short human/UI summary of what this agent does
whenToUse: Use for the precise delegation condition that should make the parent choose this agent.
tools: read, grep, find, ls, bash, contact_supervisor
thinking: high
systemPromptMode: replace
defaultExpectedOutcome: reportOnly
skills: agent-authoring
---

You are `example-agent`, an Agent Deck specialist agent.

Describe the role-specific workflow, validation expectations, and return format here.

Prompt composition

Agent Deck composes the runtime system prompt as the agent body followed by common child-session rules. The appended common rules cover completing only the assigned task, preserving parent/user decision authority, avoiding further delegation, supervisor-contact behavior, and final-result handling. Do not repeat those generic boundaries in the agent body unless the role needs a specific variation.

Keep the agent body focused on:

  • role identity and scope
  • role-specific workflow
  • validation or evidence expectations
  • output/return format
  • role-specific supervisor or progress-update cases, if any

This applies equally to bundled, global, and library/catalog agents.

Required decisions

When creating an agent, decide:

  1. Scope: user-global, library/catalog, builtin override, or builtin replacement. Ask the user before writing files when the intended scope is not explicit. Do not offer project-local agent files.
  2. Role: explorer, planner, coder, reviewer, tester, docs writer, release helper, etc.
  3. Routing: write whenToUse as one concise sentence that tells the parent exactly when to delegate to this agent; keep it distinct from the human-facing description.
  4. Tool boundary: prefer read, grep, find, ls; add bash, edit, write only when needed.
  5. Default outcome: set defaultExpectedOutcome to reportOnly, editFilesInWorktree, writeProjectFile, or directProjectWrites. Use directProjectWrites only for trusted implementation agents with edit/write; use reportOnly for explorer/planner/reviewer agents.
  6. Supervisor behavior: include contact_supervisor only when the child should ask for decisions or meaningful blockers.
  7. Continuation behavior: native subagents start fresh by default; the parent can explicitly continue a prior Subagent ID for direct follow-ups.
  8. Skills: assign explicit skill names in skills:. Agent Deck passes them through Pi native --skill injection. Do not use inheritSkills.
  9. Validation: specify what files, commands, or evidence the agent should inspect before completion.

Scope rules

Do not assume project-local scope just because the current working directory is a project. Agent Deck no longer discovers project .pi/agents or legacy project .agents as catalog sources. If the user asks to create an agent and does not specify where it should live, ask one focused question before writing:

  • user-global: ~/.pi/agent/agents/<name>.md, available across the user's projects
  • library/catalog: ~/.pi/agent/agent-library/agents/<name>.md, reusable but not automatically active until assigned

Agent Deck does not currently have a general by-reference import flow for agent markdown files like it does for skills and prompts. To bring an existing agent into Agent Deck, create or copy it into the user-global or library catalog path instead of promising an imported-by-reference agent source.

If the agent references skills, also verify the skill is visible in the global/imported skill catalog or warn before completion.

Show full SKILL.md (228 more words)Show less

Routing metadata rules

  • description is for humans and UI lists; keep it short but descriptive.
  • whenToUse is for parent-session delegation; make it a direct routing rule beginning with "Use for..." or "Use when...".
  • Keep whenToUse to one sentence and avoid repeating long descriptions.
  • Include scope/approval constraints in whenToUse when they matter, such as "approved implementation", "review-only", or "reconnaissance before planning".
  • Parent sessions use whenToUse first and fall back to description only when whenToUse is missing.

Skill rules

  • Parent sessions receive Default + current Project skill assignments.
  • Agent Deck delegated agents receive only skills explicitly listed in their skills: frontmatter or builtin override.
  • Skills are passed to Pi as native --skill <path> entries.
  • Agents with assigned skills need the read tool so Pi can load the full skill file.
  • Do not paste full skill bodies into agent prompts.

Supervisor guidance

If the agent has contact_supervisor, Agent Deck injects the generic supervisor protocol automatically. Add supervisor instructions to the agent body only for role-specific cases, such as when an explorer agent should send progress updates for discoveries that materially change the handoff.

Good defaults

  • systemPromptMode: replace for focused specialists.
  • defaultExpectedOutcome: reportOnly for research, planning, review, and advisory agents.
  • defaultExpectedOutcome: directProjectWrites for approved implementation agents that include edit/write.
  • Native delegated runs use normal project context-file discovery; do not add context-inheritance frontmatter.
  • Read-only tools for explorer/planner/reviewer; add write tools only for implementation agents.

© a-streetcoder, 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 agent-deck/bundled-skills/agent-authoring of a-streetcoder/agent-deck.

Open the folder on GitHubat commit 9efbe6c

Compare with similar skills

Agent 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.

Agent Authoring compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Agent Authoring this skilla-streetcoder/agent-deck118—~1.4kAutomated safety check: PassMIT
MCP Server Builderanthropics/skills180k64 repos~2.3kAutomated safety check: PassApache-2.0
Hook Development for Claude Code Pluginsanthropics/claude-plugins-official38k11 repos~4.1kAutomated safety check: NotesApache-2.0
Using Superpowersfarm-fe/farm5.6k35 repos~1.4kAutomated safety check: PassMIT
Executing Plans Inlineobra/superpowers296k2 repos~5.1kAutomated safety check: PassMIT
Claude Code Agent Developmentanthropics/claude-plugins-official38k8 repos~2.8kAutomated safety check: PassApache-2.0

Similar skills

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 64 repos~2.3k tokens
    Agent WorkflowsAuto-check passed

  • Hook Development for Claude Code Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to write Claude Code plugin hooks, both prompt-based checks and bash commands, for events such as PreToolUse, Stop and SessionStart.

    38k GitHub starsUsed in 11 repos~4.1k tokens
    Agent WorkflowsAuto-check: notes

  • Using Superpowers

    farm-fe/farm

    A skill your agent uses when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions

    5.6k GitHub starsUsed in 35 repos~1.4k tokens
    Agent WorkflowsAuto-check passed

  • Executing Plans Inline

    obra/superpowers

    Has the agent carry out an implementation plan itself, task by task in the current session, keeping a ledger, proving each step with a test and ending with one whole-branch review.

    296k GitHub starsUsed in 2 repos~5.1k tokens
    Agent WorkflowsAuto-check passed

  • Claude Code Agent Development

    anthropics/claude-plugins-official

    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.

    38k GitHub starsUsed in 8 repos~2.8k tokens
    Agent WorkflowsAuto-check passed

  • Skill Creator

    Azure/azqr

    Official

    Create new skills, modify and improve existing skills, and measure skill performance.

    795 GitHub starsUsed in 89 repos~8.2k tokens
    Agent WorkflowsAuto-check passed

More from a-streetcoder/agent-deck

  • MCP Install Helper

    a-streetcoder/agent-deck

    A skill your agent uses when helping users install, import, repair, or verify MCP server configurations in Agent Deck.

    118 GitHub stars~1.4k tokensUpdated 2 mo ago
    Auto-check passed

  • Loop Authoring

    a-streetcoder/agent-deck

    A skill your agent uses when creating, refining, or validating Agent Deck loops, especially to guide users iteratively through loop goal, structure, agents, write target, and safety choices.

    118 GitHub stars~1.6k tokensUpdated 2 mo ago
    Auto-check passed

  • Prompt Authoring

    a-streetcoder/agent-deck

    Create or improve Agent Deck/Pi prompt templates for reusable parent-session workflows.

    118 GitHub stars~721 tokensUpdated 2 mo ago
    Auto-check passed

  • Skill Authoring

    a-streetcoder/agent-deck

    Create or improve Agent Deck/Pi skills, including where to save them, SKILL.md frontmatter, simple vs modular structure, and validation.

    118 GitHub stars~1.8k tokensUpdated 2 mo ago
    Auto-check passed

Categories

Questions about Agent Authoring

What does Agent Authoring do?

A skill your agent uses when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment. Agent Authoring is an agent skill from a-streetcoder/agent-deck. Use when creating or reviewing Agent Deck agents, including frontmatter, tools, supervisor behavior, continuation behavior, and skill assignment.

When should I use Agent Authoring?

Agent Authoring fits situations like: reviewing Agent Deck agents; including frontmatter; supervisor behavior; continuation behavior.

How do I install Agent Authoring in Claude Code?

Run `npx skills add a-streetcoder/agent-deck --skill agent-authoring -a claude-code`. Or copy the skill folder (agent-deck/bundled-skills/agent-authoring in a-streetcoder/agent-deck) into .claude/skills/agent-authoring in your project. Claude Code loads it when a task matches its description.

How do I install Agent Authoring in Codex?

Run `npx skills add a-streetcoder/agent-deck --skill agent-authoring -a codex`. Or copy the skill folder (agent-deck/bundled-skills/agent-authoring in a-streetcoder/agent-deck) into .agents/skills/agent-authoring in your project. Codex loads it when a task matches its description.

Can I use Agent 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 a-streetcoder/agent-deck --skill agent-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/agent-authoring, .gemini/skills/agent-authoring, .github/skills/agent-authoring and .opencode/skills/agent-authoring in your project.

What does Agent Authoring need to run?

SKILL.md names no scripts, command-line tools or credentials: Agent Authoring is instructions for the agent only.

Does Agent Authoring access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Agent 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 Agent Authoring use?

Agent 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 Agent Authoring use?

About 1.4k tokens (SKILL.md is roughly 5.7k 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 Agent Authoring?

Skills that share tags, products or a category with Agent Authoring: MCP Server Builder (anthropics/skills, 180k stars), Hook Development for Claude Code Plugins (anthropics/claude-plugins-official, 38k stars), Using Superpowers (farm-fe/farm, 5.6k stars) and Executing Plans Inline (obra/superpowers, 296k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Agent Authoring?

a-streetcoder (a GitHub organization) maintains it in a-streetcoder/agent-deck, which has 118 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on August 1, 2026.

Source: a-streetcoder/agent-deck on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.