pluribus-context 0.3.16 → 0.3.18

package/CHANGELOG.md

@@ -4,6 +4,20 @@
 
 All notable changes to Pluribus are documented here.
 
+## 0.3.18 — fidelity audit discovery positioning
+
+### Changed
+
+- Reposition npm and README copy around AI-agent context fidelity audit and semantic drift detection, so package/directory discovery surfaces see the current wedge before generic context sync.
+- Track npm/GitHub discovery queries for `AI agent context fidelity audit` and `semantic drift agent context`, making the new positioning falsifiable instead of subjective.
+
+## 0.3.17 — native discovery fidelity evidence
+
+### Changed
+
+- Extend `pluribus audit --fidelity-report` with `nativeDiscoverySurface`, `resolutionAnchor`, `genericFallback`, `manualActivationRequired`, and `semanticDifference` fields so reviewers can distinguish copied files from target-native discovery and semantic preservation.
+- Document the new fidelity report fields for rule/skill bundle maintainers reviewing multi-agent installer claims.
+
 ## 0.3.16 — audit fidelity evidence
 
 ### Added

package/README.md

@@ -6,15 +6,15 @@
 [![Building in Public](https://img.shields.io/badge/building-in%20public-orange?style=flat-square)](https://x.com/RibeiroCaioCLW)
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue?style=flat-square)](LICENSE)
 
-> Intentional context across every AI tool you use.
+> Detect where AI-agent context loses fidelity across tools — then sync the parts that can be safely shared.
 
-Pluribus (`pluribus-context` on npm, `pluribus` on the command line) is an AI context sync CLI for teams and projects that use Claude Code, Cursor, GitHub Copilot, OpenClaw, Windsurf, Continue, or Zed.
+Pluribus (`pluribus-context` on npm, `pluribus` on the command line) is an AI context sync CLI with AI-agent context fidelity audit for teams and projects that use Claude Code, Cursor, GitHub Copilot, OpenClaw, Windsurf, Continue, or Zed.
 
-It keeps project instructions, conventions, constraints, and team context in one versioned source of truth, then syncs that context into the formats each AI tool expects.
+It shows where instructions keep their semantics, where they are downgraded to a generic fallback, and where manual activation or native discovery matters — then keeps project instructions, conventions, constraints, and team context in one versioned source of truth.
 
 It is **not** a persistent memory layer, retrieval system, agent orchestrator, or agent-merging framework. Think `CLAUDE.md`, `.cursorrules`, `copilot-instructions.md`, `AGENTS.md` — one intentional context, multiple generated outputs.
 
-**Reviewer shortcut:** evaluating Pluribus for a list, newsletter, package roundup, or tool directory? Use the [Community Review Packet](docs/community-review-packet.md) for copy-paste directory submission fields, safety/removability notes, feedback links, and a disposable 60-second smoke test.
+**Reviewer shortcut:** evaluating Pluribus for a list, newsletter, package roundup, or tool directory? Use the [Community Review Packet](docs/community-review-packet.md) for copy-paste directory submission fields, safety/removability notes, feedback links, and a disposable 60-second smoke test. If you only run one command, try `npx --yes pluribus-context@latest audit --json --fidelity-report` to see native discovery surfaces, generic fallbacks, manual activation requirements, and semantic differences.
 
 ---
 

package/docs/portability-fidelity-report.md

@@ -75,11 +75,23 @@ Pluribus is intentionally narrower than a skill registry or memory layer:
 - `pluribus.md` keeps the claim in one reviewed source of truth.
 - `sync --dry-run` previews target-specific outputs before writing files.
 - generated files carry a warning header so manual edits are visible.
-- `audit --json --fidelity-report` gives CI/reviewers a machine-readable check for missing/drifted outputs plus target-by-target section loss, activation shape, and portability warnings.
+- `audit --json --fidelity-report` gives CI/reviewers a machine-readable check for missing/drifted outputs plus target-by-target section loss, activation shape, native discovery surface, resolution anchor, generic fallback status, and portability warnings.
 - remote imports are opt-in, locked, cached, and digest-checked before becoming shared context.
 
 That does **not** prove runtime behavior. You still need tool-specific smoke tests for load order, path/glob activation, available tools, MCP servers, and permission semantics.
 
+### Fields to inspect in `fidelityReport.targets[]`
+
+For each selected target, the JSON report includes:
+
+- `nativeDiscoverySurface` — the file or directory pattern the target normally discovers, such as `CLAUDE.md`, `.cursorrules`, `.github/copilot-instructions.md`, or `AGENTS.md`.
+- `resolutionAnchor` — where the generated surface is resolved from today (`repo-root` for built-in targets).
+- `genericFallback` — whether the output is a broad agent fallback surface rather than a target-specific native surface.
+- `manualActivationRequired` — whether Pluribus knows the output requires manual activation after generation. Built-in project-wide targets are currently `false`; future scoped/skill targets may differ.
+- `semanticDifference` — a compact list such as `section-loss`, `project-wide-only`, or `generic-agent-file` so reviewers can distinguish “file exists” from “same behavior is preserved.”
+
+These fields are intentionally boring. They help reviewers catch cases like “installed files exist but the agent will not discover them,” or “two targets share a generic file but do not actually have the same loading semantics.”
+
 ## Suggested workflow for maintainers
 
 1. Put the portability claim in the canonical source.

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "pluribus-context",
-  "version": "0.3.16",
-  "description": "AI context/rules sync for CLAUDE.md, Claude Code, Cursor rules, Copilot instructions, OpenClaw, Windsurf, Continue, and Zed.",
+  "version": "0.3.18",
+  "description": "AI context/rules sync and fidelity audit CLI for CLAUDE.md, Claude Code, Cursor rules, Copilot instructions, OpenClaw, Windsurf, Continue, and Zed semantic drift.",
   "type": "module",
   "homepage": "https://github.com/caioribeiroclw-pixel/pluribus#readme",
   "repository": {
@@ -70,7 +70,14 @@
     "windsurf",
     "windsurf-rules",
     "zed",
-    "zed-rules"
+    "zed-rules",
+    "fidelity-audit",
+    "context-fidelity",
+    "semantic-drift",
+    "semantic-loss",
+    "native-discovery",
+    "agent-context-audit",
+    "agent-memory"
   ],
   "author": "Caio Ribeiro",
   "license": "MIT",

package/src/commands/audit.js

@@ -289,11 +289,21 @@ function buildFidelityReport({ cwd, sections, tools, loadSkill }) {
     const representedSections = new Set([...(skill?.required || []), ...(skill?.optional || [])].map((name) => name.toLowerCase()))
     const unsupportedSections = presentSections.filter((name) => !representedSections.has(name.toLowerCase()))
 
+    const outputFiles = skill?.outputFiles || []
+    const activation = inferActivation(toolId, outputFiles)
+    const discovery = inferDiscovery(toolId, outputFiles)
+    const represented = presentSections.filter((name) => representedSections.has(name.toLowerCase()))
+
     return {
       toolId,
-      files: skill?.outputFiles || [],
-      activation: inferActivation(toolId, skill?.outputFiles || []),
-      representedSections: presentSections.filter((name) => representedSections.has(name.toLowerCase())),
+      files: outputFiles,
+      nativeDiscoverySurface: discovery.nativeDiscoverySurface,
+      resolutionAnchor: discovery.resolutionAnchor,
+      genericFallback: discovery.genericFallback,
+      manualActivationRequired: discovery.manualActivationRequired,
+      activation,
+      semanticDifference: summarizeSemanticDifference({ unsupportedSections, activation, discovery }),
+      representedSections: represented,
       unsupportedSections,
     }
   })
@@ -356,6 +366,48 @@ function inferActivation(toolId, outputFiles) {
   }
 }
 
+function inferDiscovery(toolId, outputFiles) {
+  const primaryFile = outputFiles[0] || null
+  const nativeDiscoverySurfaces = {
+    claude: 'CLAUDE.md',
+    cursor: '.cursorrules',
+    openclaw: 'AGENTS.md',
+    copilot: '.github/copilot-instructions.md',
+    windsurf: '.windsurf/rules/*.md',
+    continue: '.continue/rules/*.md',
+    zed: '.rules',
+  }
+
+  return {
+    nativeDiscoverySurface: nativeDiscoverySurfaces[toolId] || primaryFile,
+    resolutionAnchor: primaryFile ? 'repo-root' : 'unknown',
+    genericFallback: toolId === 'openclaw',
+    manualActivationRequired: false,
+  }
+}
+
+function summarizeSemanticDifference({ unsupportedSections, activation, discovery }) {
+  const differences = []
+
+  if (unsupportedSections.length > 0) {
+    differences.push('section-loss')
+  }
+
+  if (activation.kind === 'flat-project-wide') {
+    differences.push('project-wide-only')
+  }
+
+  if (discovery.genericFallback) {
+    differences.push('generic-agent-file')
+  }
+
+  if (discovery.manualActivationRequired) {
+    differences.push('manual-activation-required')
+  }
+
+  return differences.length > 0 ? differences : ['no-known-template-loss']
+}
+
 function printFidelityReport(report) {
   console.log('')
   console.log('Fidelity report:')
@@ -366,7 +418,13 @@ function printFidelityReport(report) {
     const unsupported = target.unsupportedSections.length > 0
       ? `; unsupported sections: ${target.unsupportedSections.join(', ')}`
       : '; no known section loss'
-    console.log(`   • ${target.toolId}: ${target.activation.kind}${unsupported}`)
+    const discovery = target.nativeDiscoverySurface
+      ? `; surface: ${target.nativeDiscoverySurface}`
+      : ''
+    const semantics = target.semanticDifference?.length
+      ? `; semantic: ${target.semanticDifference.join(', ')}`
+      : ''
+    console.log(`   • ${target.toolId}: ${target.activation.kind}${discovery}${unsupported}${semantics}`)
   }
 
   for (const warning of report.warnings) {

package/src/utils/version.js

@@ -1 +1 @@
-export const VERSION = '0.3.16'
+export const VERSION = '0.3.18'