egm-drift-report — Claude Code skill for design drift checks
egm-drift-report — a Claude Code skill compares design documents and READMEs with the code. It runs on request to find drift or check consistency across files in a directory.
Apache-2.0 583 installs ★ 21,935 updated 2026-11-15
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a claude-codeWhat egm-drift-report does
The skill checks design documents and READMEs against the current code, then gathers discrepancies in one report. It looks for missing features and model fields, hardcoded values where the design calls for dynamic logic, and differences in architecture or libraries. The report includes discrepancies only; a document with no mismatch gets No drift detected.
Files are handled one at a time, and results are appended to {YYYY-MM-DD}_DRIFT_REPORT.md in a separate section for each document. The skill locates source files through names and symbols in the documents, or searches the code when paths are missing. Automatic invocation is disabled, so a direct request starts the check.
What it helps with
- find functions specified in a design but missing from the code
- compare model fields with the design document
- spot hardcoded values where dynamic behavior was planned
- check whether a README matches the current implementation
Example prompts for the agent
- Compare every design document in this directory with the code and report the differences.
- Check the READMEs in this directory against the current implementation.
- Find drift between the design docs and source files, then write the report.
How to install egm-drift-report
1Skills CLI (Vercel)
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a claude-code2Manually: put the files in place
~/.claude/skills/egm-drift-report/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/egm-drift-report/
1Skills CLI (Vercel)
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a codex2Manually: put the files in place
.agents/skills/egm-drift-report/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a cursor2Manually: put the files in place
.cursor/skills/egm-drift-report/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install google/elephant-goldfish-model skills/egm-drift-report2Manually: put the files in place
.github/skills/egm-drift-report/SKILL.md1Skills CLI (Vercel)
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a gemini-cli2Platform CLI
gemini skills install https://github.com/google/elephant-goldfish-model --path skills/egm-drift-reportManually: put the files in place
.devin/skills/egm-drift-report/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add google/elephant-goldfish-model --skill egm-drift-report -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about egm-drift-report
How do I install egm-drift-report in Claude Code?
Run npx skills add google/elephant-goldfish-model --skill egm-drift-report -a claude-code. The skill is installed at ~/.claude/skills/egm-drift-report/SKILL.md.
What does egm-drift-report do?
It compares design documents and READMEs with the current code, then records discrepancies in {YYYY-MM-DD}_DRIFT_REPORT.md.
Does egm-drift-report run automatically?
No. Automatic invocation is disabled; request a drift check or a consistency check across documents and code to start it.
Source text
SKILL.md · Apache-2.0
---
name: egm-drift-report
license: Apache-2.0
disable-model-invocation: true
description: >-
Runs drift detection on all design documents to check for document and
code consistency. Produces an actionable report with differences. Use when the
user asks to detect drift or check for consistency across all design documents
or READMEs in a directory.
---
# Detect Drift In Designs and Code
When running in Codex or ChatGPT Work, first read [Codex integration](../../docs/codex.md).
In Claude Code or Cowork, read `${CLAUDE_PLUGIN_ROOT}/docs/claude.md` first.
When the user asks to detect drift across all designs or a set of READMEs, the
goal is to identify all instances where the codebase has diverged or failed to
implement the specifications described by those designs or READMEs.
1. **Check Input**: Verify a directory containing design documents or READMEs
is provided. If not provided, ask the user to specify the target directory.
2. **List Documents**: List all files in the design directory.
3. **Analyze**: For EACH design document:
* **Contextualize**: Read the design document. Identify the intent,
proposed changes, and specific filenames/symbols mentioned.
* **Locate Source**: Find the corresponding source code files in the
source tree. If paths aren't explicit, use code search or grep to
locate the relevant classes, functions, or API endpoints mentioned in
the doc. Prompt the user to add missing links to the document to make
future detection cheaper.
* **Verify**: Read the current state of those source files.
* **Detect Drift**: Compare the design vs. reality. Look specifically for:
* Missing Features: Logic described in the doc but absent in the code.
* Incomplete Models: Dataclasses/Classes/Structs/Protos missing fields
specified in the design.
* Logic Shortcuts: Hardcoded values where dynamic logic was planned.
* Architectural Divergence: Implementation using different
patterns/libraries than proposed.
* **Filter**: Report ONLY the differences. Do not describe what is working
or what matches. If the code matches the design perfectly, output "No
drift detected."
4. **Report**: Produce a single report listing all discrepancies found.
* Append the analysis for each file to `{YYYY-MM-DD}_DRIFT_REPORT.md`
using your file-writing tool.
* Strict Format:
```markdown
## <DESIGN_DOC_NAME>
<Bulleted list of specific divergences found>
<Or "No drift detected">
```
5. **Guidance**
- Be rigorous. If a design doc mentions a specific object, field or file
name and the code doesn't have it, that is a reportable drift. Double
check for coverage. If there is some part of the logic that is not
referenced in a design or README, this is a reportable drift.
- Read files liberally to ensure you are looking at the current state
of the code, not relying on memory.
- Process the files sequentially to ensure the tool output buffer doesn't
overflow.
© google · Apache-2.0 · google/elephant-goldfish-model