plugin-scanner 2.0.4__tar.gz → 2.0.5__tar.gz

{plugin_scanner-2.0.4 → plugin_scanner-2.0.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: plugin-scanner
-Version: 2.0.4
+Version: 2.0.5
 Summary: Lint, verify, and gate plugin ecosystems for maintainers, CI, and publish workflows.
 Project-URL: Homepage, https://github.com/hashgraph-online/ai-plugin-scanner
 Project-URL: Repository, https://github.com/hashgraph-online/ai-plugin-scanner

{plugin_scanner-2.0.4 → plugin_scanner-2.0.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "plugin-scanner"
-version = "2.0.4"
+version = "2.0.5"
 description = "Lint, verify, and gate plugin ecosystems for maintainers, CI, and publish workflows."
 readme = "README.md"
 license = "Apache-2.0"

{plugin_scanner-2.0.4 → plugin_scanner-2.0.5}/pyproject.toml.bak

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "hol-guard"
-version = "2.0.4"
+version = "2.0.5"
 description = "Protect local AI harnesses with HOL Guard and run scanner checks for Codex, Claude, Cursor, Gemini, and OpenCode."
 readme = "README.md"
 license = "Apache-2.0"

{plugin_scanner-2.0.4 → plugin_scanner-2.0.5}/src/codex_plugin_scanner/version.py

@@ -1,3 +1,3 @@
 """Single source of truth for tool version."""
 
-__version__ = "2.0.4"
+__version__ = "2.0.5"

plugin_scanner-2.0.4/docs/guard/competitive-parity-matrix.md

@@ -1,48 +0,0 @@
-# Guard Competitive Parity Matrix
-
-## Scope
-
-This matrix compares the current Guard phase against the SafeDep-style capability set from `guard-competitive-prd.md` and `guard-competitive-todo.md`.
-
-Statuses:
-- `shipped`
-- `partial`
-- `missing`
-- `out-of-scope`
-
-Roadmap labels:
-- `parity`
-- `advantage`
-- `differentiator`
-- `later`
-
-## Capability Matrix
-
-| Capability | Status | Why | Label |
-| --- | --- | --- | --- |
-| Local install-time protection | `partial` | `guard protect` now wraps install and registration commands for package and harness flows, but zero-config harness coverage is not complete across every ecosystem path. | `parity` |
-| Runtime protection | `shipped` | Guard blocks before execution, persists receipts, carries incident context, and supports browser or terminal approval paths. | `parity` |
-| Zero-config onboarding | `partial` | Stable daemon bootstrapping and direct CLI usage work immediately, but shell shim and guided minimal-config paths are still uneven by harness. | `parity` |
-| Threat intelligence feed | `partial` | Broker verdict, advisory, revocation, and watchlist endpoints exist, but ingestion breadth and external curated feeds are still narrower than the PRD target. | `parity` |
-| Inventory and discovery | `shipped` | Local inventory, synced inventory, inventory diff, and broker inventory endpoints are implemented. | `parity` |
-| Exceptions management | `shipped` | Local and synced exceptions support owner, rationale, source, and expiry. | `parity` |
-| ABOM export | `shipped` | Guard exports local ABOM and synced broker ABOM. | `parity` |
-| Audit trail | `shipped` | Receipts, history, artifact timelines, receipt export, and local events now form a usable audit path. | `parity` |
-| GitHub App / repo-side surface | `partial` | Scanner logic can power repo-side checks, but a coherent Guard GitHub App packaging and policy bridge is still incomplete. | `parity` |
-| Centralized dashboard | `partial` | Portal and broker inventory, history, alerts, policy, and pricing surfaces exist, but frontend execution is still being refined separately. | `parity` |
-| Org policy | `shipped` | Team policy packs, synced exceptions, and alert preferences are implemented in broker and portal support layers. | `parity` |
-| Alerting | `partial` | Alert preferences, watchlist, and lifecycle events are implemented, but full outbound delivery breadth is still limited. | `parity` |
-| Pricing and funnel | `partial` | Guard pricing, entitlements, and upgrade triggers exist, but premium packaging still needs broader product-surface enforcement. | `parity` |
-| Harness-native trust | `shipped` | Guard classifies MCP servers, skills, plugin-like artifacts, commands, and harness config with harness-specific risk context. | `differentiator` |
-| Consumer-to-maintainer loop | `partial` | Receipts, advisories, trust lookups, and attestation-aware verdicts exist, but maintainer-side remediation handoff is still incomplete. | `advantage` |
-| Cross-harness trust memory | `shipped` | Guard policy, receipts, synced inventory, and team policy operate across Codex, Claude Code, Cursor, Gemini CLI, and OpenCode artifacts. | `differentiator` |
-| Maintainer-side attestation bridge | `partial` | Trust resolution and registry-attested surfaces are wired in, but deeper maintainer remediation workflows remain incomplete. | `advantage` |
-| OpenClaw remediation loop | `missing` | Consumer pain signals are not yet fully connected to maintainer-side remediation in `internet-dot`. | `later` |
-| Hermes policy consumption | `missing` | Hermes consumption of Guard policy and receipts is not yet complete in this phase. | `later` |
-
-## Summary
-
-- `parity`: install-time protection packaging, threat feed depth, GitHub app packaging, dashboard polish, and alert delivery still need more work.
-- `advantage`: consumer-to-maintainer trust loops and attestation bridges are real but not complete enough to market as finished.
-- `differentiator`: harness-native trust and cross-harness trust memory are the strongest Guard advantages today.
-- `later`: OpenClaw and Hermes integrations remain explicit later-phase work outside the repos completed here.