aaronpowell · Oct 29, 2026
@@ -66,6 +66,8 @@ ${previous}
 Recent maintainer decisions (calibrate against these as well as the guidance):
 ${formatHistory(history)}
 
+Evaluate concrete Copilot relevance and specific standalone user value across supported workflows, not only coding or text-based developer tasks. Image, video, audio, other media-generation, and creative work may be in scope where applicable; modality or non-coding use alone is not a rejection reason. Assess knowledge uplift, provenance, product/vendor-funnel evidence, responsible use, and validation appropriate to the workflow. Preserve rejection of marketing-only funnels, paid or credit-based acquisition paths without standalone user value, duplicates of native Copilot features, and generic content with no meaningful uplift.
+
 When you're done, send your result back to the session that created you (your creator) as a single message containing ONLY a JSON object with these fields: ${REVIEW_FIELD_LIST}. Include queueId: "${item.queueId}". In the rationale, explicitly address the maintainer's guidance.`;
 
     return `The maintainer requested a guided re-review of #${item.number} from the External Plugin Review Board canvas. Run it as a separate sub-session; do NOT perform the review yourself in this session.
@@ -97,6 +99,8 @@ It contains the maintainer's decision pattern, the per-issue review procedure, a
 Recent decisions made from the board (calibrate against these as well as the guidance):
 ${recent}
 
+Evaluate concrete Copilot relevance and specific standalone user value across supported workflows, not only coding or text-based developer tasks. Image, video, audio, other media-generation, and creative work may be in scope where applicable; modality or non-coding use alone is not a rejection reason. Assess knowledge uplift, provenance, product/vendor-funnel evidence, responsible use, and validation appropriate to the workflow. Preserve rejection of marketing-only funnels, paid or credit-based acquisition paths without standalone user value, duplicates of native Copilot features, and generic content with no meaningful uplift.
+
 Submissions to review:
 ${list}
 

@@ -11,7 +11,7 @@ submissions before reviewing (see "Calibrate" below).
 
 **Typically approved**
 
-- Established vendors or well-known OSS projects with genuine developer-workflow value
+- Established vendors or well-known OSS projects with genuine Copilot user value across supported workflows
   (e.g. Sumo Logic, Atlassian Teamwork Graph CLI, bitdrift, Remotion, Radius, incident.io-style SRE tooling).
 - Genuinely useful Copilot tooling and canvases that don't duplicate built-in features
   (e.g. excel-cli/excel-mcp, chat-fork-map, cache-stats, usage-insights).
@@ -29,8 +29,8 @@ submissions before reviewing (see "Calibrate" below).
 - Free skills as an on-ramp to the vendor's paid/credit-based MCP or connector: vendor LICENSE or
   `metadata.author`, `mcp add` setup commands pointing at the vendor's API, UTM/referral links, or an upstream
   README that markets the free content as a funnel — "we're not a marketing channel".
-- Crypto / payments / finance niches, or otherwise niche problem spaces — "This is not a fit for the repo."
-  / "a bit too niche a problem space for this repo."
+- Narrow submissions with no demonstrated Copilot relevance, standalone user value, or meaningful uplift —
+  "This is not a fit for the repo." / "a bit too niche a problem space for this repo."
 - Duplicates of built-in Copilot features: memory ("Copilot has a built-in memory system"), tool-use guards
   ("Copilot harnesses already have built-in provisions for controlling tool use"), markdown viewing, browser
   automation ("This can be done using Playwright, which is available in the Copilot harnesses").
@@ -40,6 +40,23 @@ submissions before reviewing (see "Calibrate" below).
 - Bulk submissions of many tiny related repos from one submitter.
 - Resubmissions: check whether the reasons for the previous rejection were actually addressed.
 
+## Assess Copilot value across workflows
+
+Do not assume submissions must address coding or other text-based developer tasks.
+Consider Copilot-supported workflows across modalities, including image, video,
+audio, other media generation, and creative work where applicable. Modality or
+non-coding use alone is not a reason to reject. Assess the concrete Copilot
+relevance, the specific user problem and standalone value, and what the plugin
+adds beyond a capable model acting without it (knowledge uplift).
+
+Keep that scope assessment separate from the quality gates: reject marketing-only
+funnels and paid or credit-based acquisition paths that provide no standalone
+user value; identify duplicates of native Copilot features and generic content
+with no meaningful uplift. Check provenance and ownership, responsible use, and
+validation evidence appropriate to the workflow (for example, meaningful output
+evaluation for a media-generation workflow rather than code-only tests). A
+non-coding workflow still needs concrete value and credible evidence.
+
 ## Procedure per issue
 
 1. `gh issue view <N> -R github/awesome-copilot --json title,body,author,comments` — read the submission

@@ -5,6 +5,15 @@ const DECISION_KINDS = ["approve", "request-changes", "comment", "close"];
 const MAX_COMMENT_LENGTH = 60000;
 const decisionsInFlight = new Set();
 
+function decisionError(error) {
+    if (/\bHTTP 422\b/.test(error?.message ?? "")) {
+        return expectedNotice(
+            "GitHub rejected this action (HTTP 422). Refresh the board and check that your account can perform this action on the pull request.",
+        );
+    }
+    return error;
+}
+
 export async function applyDecisionOnGitHub({ store, number, kind, comment, headRefOid, refresh, github = gh }) {
     if (!DECISION_KINDS.includes(kind)) throw new Error(`Unknown decision "${kind}".`);
     const key = `${store.repo}#${number}`;
@@ -26,15 +35,19 @@ export async function applyDecisionOnGitHub({ store, number, kind, comment, head
         }
 
         let result;
-        if (kind === "approve") {
-            if (!headRefOid || headRefOid !== item.headRefOid) {
-                throw expectedNotice("PR changed since it was displayed; refresh and review it again before approving.");
+        try {
+            if (kind === "approve") {
+                if (!headRefOid || headRefOid !== item.headRefOid) {
+                    throw expectedNotice("PR changed since it was displayed; refresh and review it again before approving.");
+                }
+                result = await github.approveAndEnableAutoMerge(store.repo, number, { body, headRefOid });
+            } else if (kind === "close") {
+                result = await github.closePullRequest(store.repo, number, { body });
+            } else {
+                result = await github.submitPullRequestReview(store.repo, number, { kind, body });
             }
-            result = await github.approveAndEnableAutoMerge(store.repo, number, { body, headRefOid });
-        } else if (kind === "close") {
-            result = await github.closePullRequest(store.repo, number, { body });
-        } else {
-            result = await github.submitPullRequestReview(store.repo, number, { kind, body });
+        } catch (error) {
+            throw decisionError(error);
         }
 
         const warnings = result.warning ? [result.warning] : [];

@@ -2,7 +2,7 @@ export const RECOMMENDATION_GUIDE = [
     "`approve` — the change is correct, in scope, compliant with the contribution rules, and ready to merge.",
     "`request-changes` — the idea fits the repository but concrete fixes are required before merge.",
     "`needs-insight` — a maintainer judgement call is needed (policy, duplication, product direction) that a review cannot settle.",
-    "`close` — the submission does not belong in this repository and no reasonable revision would change that (including vendor funnels and non-developer domains).",
+    "`close` — the submission does not belong in this repository and no reasonable revision would change that (including marketing-only vendor funnels and paid or credit-based acquisition paths without standalone user value).",
 ].join("\n");
 
 /** Shared schema for a single recorded review. Reused directly as the `record_review` input schema. */
@@ -23,7 +23,10 @@ export const REVIEW_FIELDS_SCHEMA = {
         },
         rationale: { type: "string", description: "One sentence explaining the deciding factor, shown on the card." },
         summary: { type: "string", description: "What the pull request actually changes." },
-        repoFit: { type: "string", description: "Repository fit and meaningful value for Copilot users." },
+        repoFit: {
+            type: "string",
+            description: "Concrete Copilot relevance and specific user value across supported workflows and modalities.",
+        },
         compliance: { type: "string", description: "Deterministic checks: frontmatter, naming, README regeneration, manifests." },
         differentiation: {
             type: "string",
@@ -67,6 +70,7 @@ function sharedRules(repo, guidancePath) {
         "  - `gh pr checks <number> -R " + repo + "` when checks are failing, to see which ones and why.",
         "  - `gh pr list -R " + repo + " --author <login> --state all` to find the author's sibling PRs. If siblings were closed for a reason, apply it consistently and cite them; flag any open sibling recorded more leniently in `checkNext`.",
         "Judge the submitted result, not assumptions about the tool that produced it. A title ending in 🤖🤖🤖 is a disclosure, not a defect.",
+        "Do not limit repository fit to coding or text-based developer tasks. Evaluate Copilot-supported workflows across modalities, including image, video, audio, other media generation, and creative work where applicable. Modality or non-coding use alone is not a rejection reason; assess concrete Copilot relevance, specific standalone user value, knowledge uplift, provenance, product/vendor-funnel evidence, responsible use, and validation appropriate to the workflow.",
         "",
         provenanceRules(),
         "",
@@ -78,7 +82,8 @@ function provenanceRules() {
     return [
         "Provenance, repo fit, and validation — these are the checks earlier reviews under-called:",
         "  - Vendor funnel: inspect the LICENSE copyright holder, frontmatter `metadata.author`, any \"Adapted from\" link, and the upstream README (`gh api -H \"Accept: application/vnd.github.raw\" repos/<owner>/<repo>/readme`). Look for MCP/connector setup commands, UTM or referral links, paid/credit upsells, and a \"free skill → paid connector\" shape, even when the vendor name was stripped from the adapted copy. A vendor funnel is `close`, not `request-changes`.",
-        "  - Repo fit: non-developer domains (growth marketing, dating, career/LinkedIn, ad creative) are not GitHub Copilot development workflows; weigh that heavily even when the content is polished.",
+        "  - Repo fit: judge concrete Copilot relevance and user value, not whether a task is coding- or text-based. Image, video, audio, media-generation, and creative workflows may be in scope where Copilot supports them; modality or non-coding use alone is not grounds to close a PR.",
+        "  - Keep repo fit separate from the other quality gates: reject marketing-only funnels, paid or credit-based acquisition paths without standalone user value, duplicates of native Copilot features, and generic content with no meaningful knowledge uplift. Check provenance, responsible use, and validation appropriate to the workflow.",
         "  - Validation: AI-only synthetic evaluations (especially fixtures built to exercise the skill's own rules) and testing only in a non-Copilot harness (e.g. Claude Code) are weak evidence. Long evaluation transcripts are not evidence of value.",
     ].join("\n");
 }

@@ -25,7 +25,7 @@ analysis into one of the board's four recommendations.
 | `approve`         | The change is correct, in scope, meets the format rules, and a maintainer could merge it as-is. Minor nits belong in the comment, not in a lower bucket.          |
 | `request-changes` | The contribution belongs in the repo but something concrete must be fixed first: missing frontmatter, bad file name, stale generated output, failing checks.      |
 | `needs-insight`   | The call is genuinely a maintainer judgement: unclear value, overlap with existing content, licensing or provenance questions, scope the reviewer cannot settle.  |
-| `close`           | The PR does not belong in this repository at all: spam, off-topic, vendor funnel, non-developer domain, duplicate of a merged PR, abandoned, or generated files only. |
+| `close`           | The PR does not belong in this repository at all: spam, off-topic, vendor funnel, duplicate of a merged PR, abandoned, or generated files only. |
 
 Default to `needs-insight` rather than guessing. A wrong `approve` or `close` costs a
 maintainer more than an honest escalation. But when the evidence shows the PR cannot be
@@ -57,9 +57,14 @@ For every pull request:
   require a particular scalar quote style.
 - **Duplication** — search the repo for an existing agent/instruction/skill covering the
   same ground. Near-duplicates are `needs-insight` unless the overlap is total.
-- **Repo fit** — non-developer domains (growth marketing, dating, career/LinkedIn, ad
-  creative, and similar) are not GitHub Copilot development workflows. Weigh that heavily
-  even when the content is polished; on its own it usually means `close`.
+- **Repo fit** — assess concrete relevance to GitHub Copilot and a specific user
+  workflow, not whether the task is coding- or text-based. Copilot-supported
+  workflows may include image, video, audio, other media generation, and creative
+  work where applicable. Modality or non-coding use alone is not a rejection
+  reason; evaluate the submission's standalone value and knowledge uplift, and
+  verify provenance, product/vendor-funnel evidence, responsible use, and
+  workflow-appropriate validation. Keep those quality checks separate from
+  assumptions about the task's modality.
 - **Knowledge uplift** — the most common way a well-formatted PR still fails. Apply the
   "Knowledge uplift" test from `.github/skills/code-review/SKILL.md` to every content
   resource. `none` is never `approve` (use `needs-insight`, or `close` if it is also a

@@ -40,8 +40,15 @@ be evaluated according to their source change.
 ## Repository fit
 
 Confirm that a submission addresses a specific GitHub Copilot workflow,
-technology, domain constraint, or user problem. Flag contributions that:
+technology, domain constraint, or user problem.
 
+Repository fit is not limited to coding or other text-based developer tasks.
+Consider the workflows Copilot supports, including image, video, audio, other
+media generation, and creative work where applicable. Modality or non-coding
+use alone is not a reason to reject; require concrete Copilot relevance and
+specific standalone user value.
+
+Flag contributions that:
 - provide generic advice that current models already handle well without
   meaningful uplift (see [Knowledge uplift](#knowledge-uplift))
 - restate an existing resource without a clear differentiator
@@ -53,6 +60,12 @@ Paid or commercial services are not automatically unsuitable. Evaluate whether
 the contribution provides standalone user value and follows the repository's
 guidance for paid-service submissions.
 
+Assess knowledge uplift, provenance, product/vendor-funnel evidence,
+responsible use, and validation appropriate to the workflow independently of
+its modality. Reject marketing-only funnels, paid or credit-based acquisition
+paths without standalone user value, duplicates of native Copilot features,
+and generic content with no meaningful uplift.
+
 ## Knowledge uplift
 
 A resource is only valuable if it changes what the model does. Well-formatted,

@@ -3,6 +3,43 @@ import { readFile } from "node:fs/promises";
 import vm from "node:vm";
 import { test } from "node:test";
 import { BoardStore } from "../.github/extensions/external-plugin-review-board/lib/state.mjs";
+import {
+    buildReviewPrompt,
+    buildRereviewPrompt,
+} from "../.github/extensions/external-plugin-review-board/lib/review-prompt.mjs";
+
+test("external-plugin review prompts and guidance assess Copilot workflows across modalities", async () => {
+    const prompts = [
+        buildReviewPrompt({ items: [], instanceId: "board-1", canvasId: "external-plugin-review-board", guidancePath: "guidance.md", history: [] }),
+        buildRereviewPrompt({
+            item: { number: 1, title: "Example", author: "someone", queueId: "q-1" },
+            instanceId: "board-1",
+            canvasId: "external-plugin-review-board",
+            guidancePath: "guidance.md",
+            history: [],
+            repo: "github/awesome-copilot",
+        }),
+    ];
+    for (const prompt of prompts) {
+        assert.match(prompt, /not only coding or text-based developer tasks/);
+        assert.match(prompt, /Image, video, audio, other media-generation, and creative work/);
+        assert.match(prompt, /modality or non-coding use alone is not a rejection reason/);
+        assert.match(prompt, /knowledge uplift, provenance, product\/vendor-funnel evidence, responsible use/);
+        assert.match(prompt, /validation appropriate to the workflow/);
+        assert.match(prompt, /paid or credit-based acquisition paths without standalone user value/);
+        assert.match(prompt, /duplicates of native Copilot features/);
+        assert.match(prompt, /generic content with no meaningful uplift/);
+    }
+
+    const guidance = await readFile(
+        new URL("../.github/extensions/external-plugin-review-board/review-guidance.md", import.meta.url),
+        "utf8",
+    );
+    assert.match(guidance, /Assess Copilot value across workflows/);
+    assert.match(guidance, /non-coding use alone is not a reason to reject/);
+    assert.match(guidance, /validation evidence appropriate to the workflow/);
+    assert.match(guidance, /duplicates of native Copilot features/);
+});
 
 test("external-plugin-review-board excludes items with pending decisions from AI review", () => {
     const store = new BoardStore("unused", "github/awesome-copilot");