Index of every page on this site: llms.txt
All skills
google avatar

/egm-review

@70406c4
by googlegoogle/elephant-goldfish-model22k stars
1,589
Run it yourself

Orchestrates the EGM Goldfish Review Protocol. Used as part of the EGM process to perform design document reviews. Executes five review phases: Initial Zoom Out Review, Comprehension Review, Design Critique, Intermediate Zoom Out Review, and Implementation Readiness.

Use this Skill: https://skilld.dev/gh/google/elephant-goldfish-model/egm-review

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

TEST.md

≈299 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Test Plan: EGM Review & Subagents

Frontmatter and link checks run in pytest tests/unit (see TESTING.md).

Protocol Verification Checklist

When evaluating or exploring egm-review interactively in chat, verify:

  • Step 1 (Intake): Gathers TargetFile and optional GoldfishContext. Before the first stage that takes GoldfishContext, adds linked files inside the working directory, asks before adding links that resolve outside it or URLs, and saves each URL in GoldfishContext as a file.
  • Step 2 (Preset Selection): Offers the four review presets (interactive, consolidated, feature spec only, implementation plan only), or maps an explicit Step argument to its corresponding preset or individual stage subflow.
  • Step 3 (Semantic Split): Discovers the split line between Feature Spec and Detailed Implementation Plan.
  • Step 5 (Initial Strategic Review): Spawns egm-zoom-out and egm-inventive-catalyst in parallel to audit altitude and false architectural compromises.
  • Subflows: Properly delegates Initial Strategic Review, Comprehension, Critic Review, Intermediate Zoom Out, and Readiness Test.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning13d3 checks · Risk SAFE
  • Gen Agent Trust Hub14d

    This skill orchestrates a multi-stage design review process through specialized subagents. It includes several security-conscious design choices, such as validating file paths against the working directory and requiring user approval before accessing external URLs. There is a potential security consideration regarding indirect prompt injection, as the skill processes external documents and URLs which could contain untrusted content. These patterns are consistent with the skill's intended purpose as a comprehensive review and remediation tool.

  • Socket13d

    No alerts

  • Snyk14d

    Risk: MEDIUM · 1 issue

Signed by skilld at 70406c4. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 3 days ago.

Activeupdated 2 weeks ago
disable-model-invocation
true

README badge

README badge for google/elephant-goldfish-model/egm-review