Files
Evan Purkhiser cc302403eb feat(build): add content-root override and shared-reference hydration (#220)
Task-shaped skills will draw their reference material (per-platform SDK
guides, per-signal concepts) from a single maintained library rather
than duplicating it. Each shipped skill must still be self-contained, so
the build copies a skill's declared references into it.

Add a CONTENT_ROOT env var (default: repo root) so the same build steps
can assemble an alternate skill tree, e.g. CONTENT_ROOT=skills-next.

hydrate-references.py walks a skills directory: for every skill with a
references.yml it brace-expands and globs the `needs:` patterns against
the library, copies the matches into the skill's references/ preserving
subpaths, and strips the manifest so it never ships. It runs via `uv run
--script` with inline PyYAML and braceexpand dependencies. build-common.sh
invokes it once per build, and only when some skill declares references,
so existing skills are unaffected and the common case spawns no
interpreter.

Factor the steps shared across all four agent builders into
build-common.sh (content-root resolution, skill copy+hydrate, commands,
SKILL_TREE), sourced by each plugin-src/<agent>/build.sh.
2026-06-26 16:52:56 -04:00
..

Sentry for Codex

The Sentry plugin for Codex. It teaches Codex how to use Sentry: SDK setup wizards for any platform, production issue debugging via the Sentry MCP server, code review with Sentry context, and monitoring configuration.

Important

This repository is generated. It is built from getsentry/sentry-for-ai and includes every skill in that library. Do not edit files here; make changes in that repository and they will be rebuilt into this one.

Install

codex plugin marketplace add getsentry/plugin-codex
codex plugin add sentry@sentry-plugin-marketplace

What's included

  • The full Sentry skill library (SDK setup wizards, debugging and code-review workflows, feature setup).
  • The hosted Sentry MCP server for querying your Sentry environment.

A few entry-point skills are always available; the rest are hidden behind them using Codex's agents/openai.yaml (policy.allow_implicit_invocation: false) and surface on demand, so they do not crowd the model's context.