heygen-cli: Skill for Claude Code

.agents/skills/release-cli/SKILL.md

release-cli is a skill for Claude Code, Codex from heygen-com/heygen-cli. It costs 115 tokens per session (1,898 once invoked), scanned A, original, Apache-2.0.

A guided release workflow for the heygen command-line tool. It follows the project's release checklist, prepares the version and changelog, runs end-to-end tests, starts the release workflow, publishes notes, and checks the installer while asking for approval at defined decision points.

In plain words
What is it for?
Use it to prepare stable or development releases, land pending pull requests, verify the command surface, run end-to-end tests, publish release notes, and confirm that installation works.
Why use it?
Releasing software involves many ordered steps and potentially irreversible actions such as merging pull requests or publishing a version. This workflow adds checks, uses a fresh copy of the main branch, and stops for explicit approval before those actions.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is heygen-com/heygen-cli's own configuration. It tells Claude Code and Codex how to work on heygen-cli itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything heygen-cli configures →

Reuse

Borrowing it

Nothing to install: this file belongs to heygen-com/heygen-cli. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/heygen-com/heygen-cli/main/.agents/skills/release-cli/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/heygen-com/heygen-cli

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for release-cli

README.md
[![agentmods](https://agentmods.dev/badge/skills/heygen-com/heygen-cli/release-cli/github.svg)](https://agentmods.dev/skills/heygen-com/heygen-cli/release-cli)
Your own site
<a href="https://agentmods.dev/skills/heygen-com/heygen-cli/release-cli"><img src="https://agentmods.dev/badge/skills/heygen-com/heygen-cli/release-cli/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for release-cli

Your own site · 80×15
<a href="https://agentmods.dev/skills/heygen-com/heygen-cli/release-cli"><img src="https://agentmods.dev/badge/skills/heygen-com/heygen-cli/release-cli.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,898 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00115 $0.01898
Opus 5.5 $0.00046 $0.00759
Sonnet 5.5 $0.00023 $0.00380
Haiku 4.5 $0.00012 $0.00190

Measured 21d ago against content hash 06dd77ab7401, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

release-cli scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 21d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

`install.sh` reads `INSTALL_DIR`, so it goes on `bash`, not `curl`: `d=$(mktemp -d); curl -fsSL https://static.heygen.ai/cli/install.sh | INSTALL_DIR="$d" bash`,
.agents/skills/release-cli/SKILL.md · 146 lines

How it starts

The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Release the heygen CLI

RELEASE.md is the source of truth for every step below; this skill runs it in order and adds where to stop and what tends to go wrong. When the two disagree, RELEASE.md wins, and fix this file.

Work from a fresh fetch, never from the local main checkout, which may be behind. Set LAST_STABLE and RELEASE_COMMIT exactly as RELEASE.md's pre-release checklist shows and keep them for the whole run: RELEASE_COMMIT is what every step checks and what stop 4 releases. If a merge at step 2 moves origin/main, fetch, set RELEASE_COMMIT again, and redo from step 1.

Stop points

Stop and ask the user before each of these, with what you found and a recommendation. Never proceed on a guess. Each stop needs an explicit yes for that specific action, in this session: the original request, a yes at an earlier stop, or silence does not count. The one exception is stop 4, whose yes also covers publishing its notes. If the action changes after the yes (a new head after a rebase, a different commit on main, edited notes), ask again.

  1. Changing any open PR before the release (step 2): pushing to it (examples for a sync PR, a rebase onto main), approving it, or merging it.
  2. Running E2E (step 5): it spends API credits and needs HEYGEN_API_KEY. Ask which key to use; do not go looking for one.
  3. The version (step 6), whenever step 4 found anything breaking or you are unsure between patch and minor.
  4. Triggering the release workflow. Show the version, RELEASE_COMMIT, and the notes file step 9 will publish (a dev release has none), and record its shasum -a 256. Dispatch with -f commit=<that SHA> so the workflow releases exactly that commit. A yes here also covers publishing that file at step 9 for this version.
  5. Replacing the published release notes (step 9), unless re-running shasum -a 256 on the file then matches the hash recorded at the most recent stop 4 yes in this session for this version: show the final notes. A lost hash means asking.
  6. Any recovery action (step 8) that changes the repository or a release: deleting a tag or release, dispatching again, or re-running a job. Say which job failed, what RELEASE.md prescribes, and what the action will change.

Read the full file on GitHub · 146 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 21d ago Changed · +6 lines 06dd77ab7401
  2. 22d ago First seen · 140 lines · 115 tokens per session scan A 2839ad0e890a

Subscribe to this mod's changes

release-cli is a skill published in the GitHub repository heygen-com/heygen-cli (164 stars, last pushed today), licensed Apache-2.0. It adds 115 tokens to every session and 1,898 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-08.