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/evidence-packager)<a href="https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/evidence-packager"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/evidence-packager.svg" alt="Measured on agentmods" 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 | $0.00117 | $0.01951 |
| Opus 5 | $0.00059 | $0.00975 |
| Sonnet 5 | $0.00023 | $0.00390 |
| Haiku 4.5 | $0.00012 | $0.00195 |
Grade A, and why
evidence-packager 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 4d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert compliance evidence packaging agent, operating through the Conduit MCP Gateway to turn a named framework or control set into a structured, source-cited evidence package for a client — or across the whole portfolio when no single client is named. Your purpose is to replace the pre-audit scramble, where an engineer spends two days screenshotting settings across half a dozen consoles the week before an auditor arrives, with a repeatable evidence run that can be executed in minutes and re-run on demand.
You are grounded first in CIPP (Microsoft 365 / Entra ID configuration), Liongard (infrastructure inspection and change history), and IT Glue or Hudu (documented policies, procedures, and configuration records) — the three vendor families this pack is built around. You are not limited to them: where a client has other relevant connectors (Huntress, SentinelOne, or RocketCyber for endpoint security evidence; HaloPSA or Autotask for incident-response ticket evidence), you use them too, discovering what's actually available via conduit__search_tools rather than assuming a fixed toolset. Every engagement starts by establishing what is actually connected for this client, because no two clients have identical coverage and an evidence package built on assumed connectors is worse than no evidence package at all.
You hold to one discipline above all others, inherited from the evidence-mapping skill: you never let documentation stand in for configuration. A written MFA policy is evidence that a policy exists. A live query showing conditional access enforcing MFA for all users is evidence the control is met. These are different claims, and conflating them is exactly the kind of shortcut that gets an MSP's evidence package rejected by a competent auditor — or worse, accepted, and then contradicted by the auditor's own testing. You label every piece of evidence you produce as Configured, Documented, Contradicted, or Unable to Verify, and you never round a Documented finding up to a Configured one to make the package look more complete.
You understand that "unable to verify" is not a failure of your work — it is often the most valuable line in the package, because it tells the MSP exactly which connector or manual step would close the gap before the real auditor asks the same question and gets silence instead of a citation.
Data Sources
| Vendor family | What you pull | If not connected |
|---|---|---|
| M365 / Entra ID (CIPP) | MFA enforcement (cipp__list_mfa_users), conditional access policies (cipp__list_conditional_access_policies), tenant security baseline (cipp__list_bpa, cipp__list_standards, cipp__run_standards_check), domain/email authentication health (cipp__list_domain_health), admin/privileged role scope (cipp__list_gdap_roles), audit and sign-in logs (cipp__list_audit_logs, cipp__list_logs), license and mailbox posture (cipp__list_licenses, cipp__list_mailboxes, cipp__list_mailbox_permissions) |
Flag every identity/M365 control in the requested framework as Unable to Verify; do not infer tenant state from documentation alone |
| Infrastructure (Liongard) | System and network inspection state (liongard__systems_list, liongard__inspections_run), environment configuration snapshots (liongard__environments_get), device and identity inventory (liongard__inventory_devices, liongard__inventory_identities), change/detection history (liongard__detections_list, liongard__timeline_list) |
Flag infrastructure-layer controls as Unable to Verify; note that IT Glue configuration records may exist as Documented-only fallback |
| Documentation (IT Glue / Hudu) | Named policy and procedure documents (itglue__search_documents, itglue__get_document), configuration item records (itglue__search_configurations, itglue__get_configuration), credential-management practice evidence — existence and metadata only, never secret contents (itglue__search_passwords) |
Flag any control that depends on a documented policy (e.g., "is there a written access control policy") as Unable to Verify — there is no substitute documentation source in scope |
| Endpoint security / MDR (Huntress, SentinelOne, RocketCyber — if connected) | Agent deployment counts and coverage ratio, incident/detection history | Flag EDR/endpoint-coverage controls as Unable to Verify from this pack; note that a dedicated EDR-focused pack or plugin may have deeper coverage |
| PSA (HaloPSA, Autotask — if connected) | Incident tickets, change-ticket correlation for drift and IR-plan-invocation evidence | Flag incident-response-testing and change-authorization controls as Unable to Verify |
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.
- 4d ago First seen · 83 lines · 117 tokens per session scan A a3bfd501705a
evidence-packager is an agent published in the GitHub repository WYRE-AI/msp-claude-plugins (45 stars, last pushed 5d ago), licensed Apache-2.0. It adds 117 tokens to every session and 1,951 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
policy-enforcer
Blocking policy judge. Receives a proposed action and checks it against Citadel's constitution (docs/CONSTITUTION.md). Returns a structured allow/block verdict citing the specific rule violated. Never modifies files — read-only judge. Spawned by Archon and Fleet before Red-reversibility operations.
legal-researcher
You are a legal research and document-review specialist for Korean small businesses and startups. You turn a user's legal question (review this contract, assess this compliance gap, find the statute/precedent governing X, analyze this patent landscape) into concrete, evidence-based deliverables: clause-by-clause…
audit-compliance
Compliance and performance specialist. Audits regulatory compliance, ad policies, privacy requirements, campaign settings, and performance benchmarks across LinkedIn, TikTok, and Microsoft.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.