Agent Skills for building secure applications with the Listen Notes Podcast API.
The podcast-api skill helps coding agents inspect an existing project, select the right endpoint, keep credentials server-side, implement pagination and error handling, test against the mock server, and answer API or billing questions from authoritative sources.
Questions? Contact us at hello@listennotes.com.
npx skills add PodcastAPI/skillsFrom Codex, use the built-in skill installer and ask it to install this repository's podcast-api skill:
$skill-installer
Install the podcast-api skill from https://github.com/PodcastAPI/skills/tree/main/skills/podcast-api
Or install it non-interactively with the Skills CLI:
npx skills add https://github.com/PodcastAPI/skills \
--skill podcast-api \
--agent codex \
--global \
--yesFor a manual user-wide installation, copy the skill directory to $HOME/.agents/skills/podcast-api. Codex detects skill changes automatically; restart Codex if the skill does not appear. Run /skills or mention $podcast-api in a prompt to invoke it explicitly.
See the official Codex Agent Skills documentation.
/plugin marketplace add podcastapi/skills
/plugin install podcast-api@podcastapi
Clone the repository and copy skills/podcast-api into the appropriate global skill directory:
| Agent | Global skill directory |
|---|---|
| OpenAI Codex | $HOME/.agents/skills/ |
| Claude Code | ~/.claude/skills/ |
| Cursor | ~/.cursor/skills/ |
Compatible agents can install the same folder in their documented Agent Skills directory.
The skill keeps procedural guidance local while treating these live resources as authoritative:
- OpenAPI YAML for endpoint contracts
- API tutorials for application workflows
- API FAQ and billing FAQ for common questions
- Pricing and terms for current plan and policy details
Exact request and response schemas are intentionally not copied into this repository.
CI validates the skill against the Agent Skills specification, checks documentation links, and smoke-tests installation for Codex, Claude Code, and Cursor. Representative prompts and expected safety invariants live in evals/cases.json; they require no production API key or model secret.