Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add basefoundry/base --skill base-release-traingit clone --depth 1 https://github.com/basefoundry/baseWrote 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.
[](https://agentmods.dev/skills/basefoundry/base/base-release-train)<a href="https://agentmods.dev/skills/basefoundry/base/base-release-train"><img src="https://agentmods.dev/badge/skills/basefoundry/base/base-release-train/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.
<a href="https://agentmods.dev/skills/basefoundry/base/base-release-train"><img src="https://agentmods.dev/badge/skills/basefoundry/base/base-release-train.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.00559 |
| Opus 5.5 | $0.00014 | $0.00224 |
| Sonnet 5.5 | $0.00007 | $0.00112 |
| Haiku 4.5 | $0.00003 | $0.00056 |
Grade A, and why
base-release-train scanned grade A with 0 findings 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 4d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Base Release Train
Use this skill for Base, base-demo, base-cli, base-bash-libs, and the two demo repositories when the task concerns release readiness, release milestones, cross-repository pins, BOM evidence, or publication follow-up.
Source of truth
Read docs/release-train.md and .release/release-train.json in the Base
checkout before making a recommendation. Use
python3 tools/release_train_check.py --format json for deterministic local
preflight and add --remote for GitHub milestone/release readback.
Report separately:
- completed and publicly verified evidence;
- blockers that must be resolved before publication;
- warnings or stale issues that need triage;
- intentionally deferred work assigned to a later milestone.
Do not infer a release from main, a README version, green ordinary CI, or a
mutable branch. Require the public tag, peeled full commit, release assets,
checksums, required BOM platforms, and post-publication readback.
Operating modes
- assess: read current repositories, milestones, pull requests, hosted checks, tags/releases, pins, lockfiles, docs, BOMs, and public metadata.
- plan: create or reconcile exact SemVer milestones and issue ledgers; default a missing future companion milestone to the next patch only when the issue is explicitly release work.
- check: run the local and hosted preflight and preserve machine-readable output in the release issue.
- train: implement issue-backed changes in canonical worktrees, validate, push, and open reviewable PRs. Keep the exact issue ledger.
- readback: verify tags, releases, assets, registries, checksums, immutable metadata, BOM contents, and public docs after publication.
Safety boundaries
The skill is read-only by default. Before creating or editing issues, milestones, Project fields, branches, PRs, releases, or Homebrew state, obtain explicit authorization for that mutation. Never merge or publish just because checks are green. Treat queued checks, unresolved review threads, stale provenance, missing milestones, and non-immutable releases as blockers.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago First seen · 56 lines · 35 tokens per session scan A 82eabde43aa9
base-release-train is a skill published in the GitHub repository basefoundry/base (120 stars, last pushed yesterday), licensed Apache-2.0. It adds 35 tokens to every session and 559 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-12.
Other skills, from other repositories
github
Drive GitHub via the official gh CLI — repos, issues, pull requests, releases, gists, Actions runs, and raw REST through gh api. Use when the user asks to inspect or manage GitHub.
github
GitHub CLI (gh) for repository management, issues, pull requests, releases, workflows, search, gists, and API access. Use when the user asks about any GitHub operation — creating repos, forking, cloning, PR workflows, issue triage, CI/CD runs, code search, release management, or direct API calls. Keywords: github, gh…
pi-review-gated-delivery
Use when coordinating review-gated delivery of discrete GitHub issues between one implementation peer and a review-only lead, with user-owned Changesets and approval-gated commits.
release
Prepare a new release by updating CHANGELOG.md, verifying documentation (README.md, CLAUDE.md, AGENTS.md, ROADMAP.md, docs/), bumping patch version in package.json, building, and suggesting publish commands. Use when asked to release, bump version, prepare release, or publish.
4x-autopilot
An owner-only automation mode for the 4x feature pipeline, which manages tracked development tasks from selection through completion and merging. It runs one feature at a time and reads project settings to determine the repositories and workflow.
github-repo-management
Clone, create, fork, configure, and manage GitHub repositories. Manage remotes, secrets, releases, and workflows. Works with gh CLI or falls back to git + GitHub REST API via curl.