aftergrid
v0.0.0Open-source skills, Checks and a thin CLI for producing Findings: analysis memos a non-data Reader can understand, inspect and act on, with every number traced to its evidence.
By Patrick Morris0 GitHub starsUpdated 6 days ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 9 skill or MCP entries
- Source updated
- Sep 17, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install aftergrid for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install aftergrid@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/patrickjmorris/aftergridClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The repository root is the plugin root.
Plugin files
├── .claude-plugin/plugin.json├── skills/analysis-review/SKILL.md├── skills/analyze/SKILL.md├── skills/checked-analysis/SKILL.md├── skills/grill-question/SKILL.md├── skills/iterate-visual/SKILL.md├── skills/revise-finding/SKILL.md├── skills/setup-aftergrid/SKILL.md├── skills/shape-narrative/SKILL.md└── skills/write-finding/SKILL.md
Included Skills9
Review a complete Finding with three independent reviewers — Method, Question and Reader — and record what each found in the manifest. Use after the draft is written and before `aftergrid check`, or when a revision has left the recorded reviews stale.
Take a raw ask, or an already sharpened Question, all the way to a reviewed Finding draft — clarify, analyse under Checks, write, illustrate, shape, review, check. Use to run an analysis end to end, or to resume one that stopped for missing input, a blocking review or a harness that refused a write.
Run one Analysis and leave an Analysis directory the writer can consume — probe the catalog, write the Checks before the analysis SQL, run each query and Check through the harness's own data tool and record what ran, fill analysis.yaml. Where the Instance configures an adapter, capture the inputs and execute against them instead. Use after a Question is sharpened, when an analysis must rerun on the same evidence, or when a run must stop on a missing definition approval, clarification or provisional sign-off.
Grill a raw ask until it is a Question — decision, Reader, metric, population, window, primary comparison and a falsifier — and write it into a Finding. Use before any analysis runs, when an ask names no metric or window, when a definition is missing or only proposed, or to resume a Question left unresolved.
Render a Finding's charts, look at each one and revise it against the Storytelling with Data rubric. Use after writing or changing a chart spec, before a Finding goes to a Reader, or when a chart is hard to read.
Take an Operator's feedback on a reviewed Finding, classify it as presentation, interpretation or numeric, and apply it as a new revision or say what it reopens. Use for "make it horizontal", "lead with retention", "reword the answer", or any edit to a Finding that has already been checked.
Set up an aftergrid Instance in this repository — scaffold the layout, check the dependencies, validate the warehouse connection when the team has asked for one, install the guardrail hook and report what is still missing. An adapter is optional; the recorded path is the default.
Shape a written Finding so a Reader gets the point without reading all of it - answer first, one Claim per Evidence subsection with the heading stating the Claim, chart titles stating the Claim, the material caveat beside the Answer, and the profile's words rather than the analyst's. Use after /write-finding and /iterate-visual, before /analysis-review.
Turn a checked Analysis directory into a Finding a Reader can read and inspect - memo.md in the fixed sections, typed Claims, charts and tables, every displayed value bound to a saved result. Use once /checked-analysis has pinned results and written analysis.yaml, and before /iterate-visual, /shape-narrative and /analysis-review.
Plugin manifests1
{
"name": "aftergrid",
"description": "Open-source skills, Checks and a thin CLI for producing Findings: analysis memos a non-data Reader can understand, inspect and act on, with every number traced to its evidence.",
"version": "0.0.0",
"author": {
"name": "Patrick Morris",
"url": "https://github.com/patrickjmorris/aftergrid"
},
"skills": [
"./skills/analysis-review",
"./skills/analyze",
"./skills/checked-analysis",
"./skills/grill-question",
"./skills/iterate-visual",
"./skills/revise-finding",
"./skills/setup-aftergrid",
"./skills/shape-narrative",
"./skills/write-finding"
]
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[aftergrid on Agent Plugins Marketplace](https://pluginsmp.com/plugins/aftergrid)