# Agent Card

> Project processkit agent-card Artifacts into canonical public agent-card JSON files. Use when publishing an agent identity, endpoint, capability, or interoperability card from repository context into a runtime-visible file.

- Skill: `projectious-work/agent-card` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add projectious-work/agent-card`
- Raw SKILL.md: https://api.skillmd.com/api/skills/projectious-work/agent-card/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: projectious-work (https://skillmd.com/u/projectious-work)
- Updated: 2026-08-19
- Page: https://skillmd.com/skills/projectious-work/agent-card

---


# Agent Card

## Intro

Agent cards describe a runnable or externally visible agent in a
provider-neutral JSON shape. Store the canonical source as an Artifact
with `spec.kind=agent-card`; use `project_agent_card` to render the
public JSON file and return a checksum.

The source Artifact may put the card under `spec.card`, or it may store
YAML/JSON card content in the Markdown body. Optional Actor data can
augment the projected card with runtime interfaces.

## Overview

`project_agent_card` reads an agent-card Artifact, validates the card
payload, writes the projected JSON file, and records projection metadata
such as path and checksum on the source Artifact. The projection is a
runtime file; the Artifact remains the reviewable source of truth.

## Gotchas

- Do not edit projected card JSON as the canonical source. Update the
  Artifact and re-project.
- Do not put secrets in public card payloads. Use private configuration
  outside the projected interoperability file.
- Re-project after endpoint, capability, or identity changes so the
  checksum stays useful.

## Full reference

### MCP tools

- `project_agent_card`

### Source entity

Use `Artifact(spec.kind=agent-card)` with card content in `spec.card`
or the Markdown body.