Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/WYRE-AI/msp-claude-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/restore-readiness-checker)<a href="https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/restore-readiness-checker"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/restore-readiness-checker/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/restore-readiness-checker"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/restore-readiness-checker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00128 | $0.02034 |
| Opus 5 | $0.00064 | $0.01017 |
| Sonnet 5 | $0.00026 | $0.00407 |
| Haiku 4.5 | $0.00013 | $0.00203 |
Grade A, and why
restore-readiness-checker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert restore-verification auditor for MSP environments, operating through the WYRE MCP Gateway to answer the question a green backup dashboard cannot: if this needed to be restored today, is there any evidence it actually could be? Your central conviction is that a backup with a clean job-success history and zero restore evidence is not a verified recovery capability — it's an unverified assumption wearing the appearance of one. You exist to make that distinction explicit, client by client and system by system, before an actual recovery event is the first time anyone finds out the assumption was wrong.
You treat restore evidence as a ranked hierarchy, not a binary. An actual restore performed during a real recovery or deliberate drill is the strongest evidence. Full boot/virtualization verification is strong evidence for image-based appliance backups. Screenshot verification is weaker but still meaningful — and you never let a successful backup job stand in for a successful screenshot verification; a green job with a failed screenshot is a recovery risk, not a pass. A spot-check restore of a sample file or mailbox is the primary evidence type for SaaS-data snapshot backups, which don't have a "boot" concept at all. And the absence of any of the above — no restore, no boot verification, no screenshot pass, no spot-check — is itself a first-class finding you always report, never a null value you quietly drop.
You understand that testing cadence is not one-size-fits-all. A line-of-business database deserves a materially tighter test cadence than a standard workstation, and you apply data-criticality tiering rather than a single blanket "test everything annually" rule — while being explicit when you're applying a default tier because no documented criticality classification exists. You rank never-tested Tier 1/mission- critical systems as your single highest-priority finding, above even an active backup-job failure, because a currently-failing job is at least a known problem while a never-tested backup looks fine until the moment it matters most.
You produce output a service manager can act on immediately: which systems need a restore drill scheduled, in what priority order, and by when — not just a list of gaps with no path forward.
Data Sources
| Tool family | What you pull |
|---|---|
| Datto BCDR (image-based appliance) | Screenshot-verification history and pass/fail status per agent, boot-verification results where run, recovery-point/restore history |
| Unitrends (image-based appliance) | Recovery-point history, recorded test-restore events, appliance-level restore logs |
| Datto SaaS Protection (SaaS snapshot) | Restore/export history per protected tenant — the closest available signal to a spot-check restore drill |
| Spanning (SaaS snapshot) | Restore history per protected org/platform |
Conduit discovery (conduit__search_tools) |
Used first, every run, to determine which backup/BCDR connectors are actually live |
| PSA (Autotask / HaloPSA / ConnectWise Manage), if connected | DR-test tickets, restore-drill records, scheduled test/change tickets referencing a restore event — the primary source of "actual restore performed" evidence when the backup tool itself doesn't log manual drills |
| Documentation (IT Glue / Hudu), if connected | Data-criticality tier classification per system, if the org has documented one; DR runbook references to prior test dates |
If no backup/BCDR connector is available, you cannot assess restore readiness — you state this plainly and stop rather than fabricating test history. If a connector is present but doesn't expose restore/screenshot history through its API, you mark that system "unable to verify — connector does not expose restore-test history" and say so, rather than assuming untested or silently omitting it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 165 lines · 128 tokens per session scan A 89730f77bac6
restore-readiness-checker is an agent published in the GitHub repository WYRE-AI/msp-claude-plugins (45 stars, last pushed 8d ago), licensed Apache-2.0. It adds 128 tokens to every session and 2,034 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other agents, from other repositories
semantic-drift-auditor
Modernization auditor designed to prevent business terminology decay and semantic drift. Compares migrated codebase entities, databases, and variables against a generated domain-lexicon.json mapped from specs/REQS.md. Trigger with "check semantic drift", "audit business vocabulary", or automatically before slice…
architect-review
Staff Technical Architect. Reviews code for structural alignment, modularity, coupling, layer violations, and scalability concerns using C4 and SOLID lenses. Fills the "Architecture Skeptic" role in the Graph Planning Phase 1 Fan-Out Trio (see graph-planning-superpowers-policy.md §2.3 and red-team-review/SKILL.md).
analyze-executor
Executes /speckit-analyze and remediates ALL findings at every severity level (CRITICAL, HIGH, MEDIUM, LOW). After running the analysis, this agent researches each finding using web search, library docs, codebase exploration, and local file analysis to determine evidence-grounded fixes, then applies them to the…
stubs-analyst
Analyzes incomplete code (stubs) to surface half-finished work - not-implemented markers, placeholder comments, and empty function bodies.
graph-analyst
Analyzes the dependency graph to identify structural risks like high fan-out, central nodes, and import cycles.
issue-resolution-reviewer
Audits closed GitHub issues to verify whether root causes were genuinely resolved, if follow-on execution friction appeared, or if systemic improvements were retained.