How do I install the Aster Review CI skill?

Run npx skillmds@latest add zfinix/aster-review-ci in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Aster Review CI safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Aster Review CI?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Aster Review CI free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Aster Review CI?

zfinix (@zfinix) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Aster Review CI by zfinix — Agent Skill

Name: Aster Review CI
Author: zfinix

Aster Review CI

Run aster code reviews non-interactively in CI, GitHub Actions, or from another agent. Covers `aster review --pr`, `--json`, `--stream`, `--comment`, diff-from-stdin, token handling, and filtering findings. Use when wiring aster into a pipeline, posting PR comments, or parsing review output programmatically.

zfinix Updated Sep 17, 2026

File contents

Aster review in CI and automation

aster review reads a diff, hypothesizes defects, verifies them against the indexed repo, and reports findings. Everything is flag-driven; nothing requires a terminal.

Picking the diff

aster review                          # current branch vs its base
aster review --range main..HEAD       # explicit git range
aster review --diff changes.patch     # a saved diff file
git diff main... | aster review --diff -   # diff from stdin
aster review --pr 123                 # fetch a GitHub PR's diff

--pr needs a token: --token, else GITHUB_TOKEN, else the one stored by aster login. The repo defaults to the origin remote; override with --repo owner/repo.

Machine-readable output

aster review --json                   # findings as one JSON array
aster review --stream                 # NDJSON progress events, one per line

Use --json in CI and pipe it to aster fix or your own tooling. --stream is for editors and UIs that want live progress. --tui is the interactive browser; never use it headless.

Posting to the PR

aster review --pr 123 --comment       # post findings as inline PR comments

--comment implies --pr. Run it only when you intend to publish; the plain run is read-only.

Scoping and thresholds

aster review -i "src/**/*.rs" -x "**/generated/**" --min-confidence 0.7

-i/--include overrides include from aster.yaml; -x/--exclude adds to it. Both repeatable.
--min-confidence <0.0-1.0> drops weak findings; falls back to aster.yaml.
--no-index skips the symbol index for speed, at the cost of weaker verification. Prefer keeping the index in CI.
--repo-root <DIR> points evidence retrieval at a checkout other than the cwd.

Minimal GitHub Actions shape

- run: aster review --pr ${{ github.event.pull_request.number }} --comment --json > findings.json
  env:
    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
    ASTER_API_KEY: ${{ secrets.ASTER_API_KEY }}

The API key comes from ASTER_API_KEY (never from aster.yaml). An aster.yaml committed to the repo supplies models, analyzers, and filters; aster init -y writes a default one.

---name: aster-review-cidescription: Run aster code reviews non-interactively in CI, GitHub Actions, or from another agent. Covers `aster review --pr`, `--json`, `--stream`, `--comment`, diff-from-stdin, token handling, and filtering findings. Use when wiring aster into a pipeline, posting PR comments, or parsing review output programmatically.---# Aster review in CI and automation`aster review` reads a diff, hypothesizes defects, verifies them against the indexed repo, and reports findings. Everything is flag-driven; nothing requires a terminal.## Picking the diff```shaster review # current branch vs its baseaster review --range main..HEAD # explicit git rangeaster review --diff changes.patch # a saved diff filegit diff main... | aster review --diff - # diff from stdinaster review --pr 123 # fetch a GitHub PR's diff````--pr` needs a token: `--token`, else `GITHUB_TOKEN`, else the one stored by `aster login`. The repo defaults to the `origin` remote; override with `--repo owner/repo`.## Machine-readable output```shaster review --json # findings as one JSON arrayaster review --stream # NDJSON progress events, one per line```Use `--json` in CI and pipe it to `aster fix` or your own tooling. `--stream` is for editors and UIs that want live progress. `--tui` is the interactive browser; never use it headless.## Posting to the PR```shaster review --pr 123 --comment # post findings as inline PR comments````--comment` implies `--pr`. Run it only when you intend to publish; the plain run is read-only.## Scoping and thresholds```shaster review -i "src/**/*.rs" -x "**/generated/**" --min-confidence 0.7```- `-i/--include` overrides `include` from aster.yaml; `-x/--exclude` adds to it. Both repeatable.- `--min-confidence <0.0-1.0>` drops weak findings; falls back to aster.yaml.- `--no-index` skips the symbol index for speed, at the cost of weaker verification. Prefer keeping the index in CI.- `--repo-root <DIR>` points evidence retrieval at a checkout other than the cwd.## Minimal GitHub Actions shape```yaml- run: aster review --pr ${{ github.event.pull_request.number }} --comment --json > findings.json env: GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} ASTER_API_KEY: ${{ secrets.ASTER_API_KEY }}```The API key comes from `ASTER_API_KEY` (never from aster.yaml). An `aster.yaml` committed to the repo supplies models, analyzers, and filters; `aster init -y` writes a default one.

Frequently asked questions