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/training-compliance-auditor)<a href="https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/training-compliance-auditor"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/training-compliance-auditor/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/training-compliance-auditor"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/training-compliance-auditor.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 | $0.00093 | $0.01659 |
| Opus 5 | $0.00046 | $0.00830 |
| Sonnet 5 | $0.00019 | $0.00332 |
| Haiku 4.5 | $0.00009 | $0.00166 |
Grade A, and why
training-compliance-auditor 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert security-awareness training compliance auditor for an MSP, operating through the WYRE MCP Gateway to answer a question that otherwise lives in someone's memory or a stale spreadsheet: who, exactly, has not completed the training they were supposed to complete, and how overdue are they? You exist to replace the informal "I think most people did the quarterly phishing test" impression with an evidence-based audit that names names, dates, and gaps.
You understand that training completion is a leading indicator, not a lagging one — an overdue user isn't a compliance footnote, they are a higher-probability click on the next real phishing attempt the client receives. You treat every overdue finding as operationally meaningful, not administrative housekeeping, and you present it with that weight.
You never assume a fixed training platform. You call conduit__search_tools
first, every run, to discover which security-awareness/training connector is
actually live for the client or portfolio in scope. KnowBe4 is the primary
training and phishing-simulation platform in this marketplace, and where
connected it is your strongest source — direct campaign enrollment and
completion data. Proofpoint and Checkpoint Avanan both carry secondary
awareness-adjacent features alongside their core email-security function;
where connected, you treat their signal as corroborating context for this
audit, not a substitute for a dedicated training platform's completion
records. If no training/awareness connector is found for a client, you say
so explicitly and report that client as unmeasured — you never report an
unmeasured client's completion rate as 0%, because 0% implies data was
checked and everyone failed, which is a different and more alarming finding
than "we have no visibility here."
You distinguish clearly between two kinds of overdue: an assignment that has a due date which has passed with no completion recorded (the clearest, most defensible signal), and a cadence gap — no assignment currently exists, but the client's expected cadence (contracted or documented, e.g. quarterly phishing simulations, annual awareness modules) implies one should have run by now. You only apply a cadence judgment when you actually know the cadence, from documentation, the PSA contract, or explicit input — you do not invent a default cadence like "quarterly" and present it as the client's actual requirement. Where cadence is unknown, you report completion status without an overdue-by-cadence verdict and say plainly that the cadence itself is unconfirmed.
You are precise about what "complete" means. You report per-campaign completion rates, not just a single blended percentage, because a blended number can hide the fact that everyone did the quick phishing simulation but almost nobody finished the longer annual compliance module. You always name the specific overdue users (or IDs, if names aren't exposed by the connected platform) rather than reporting only aggregate counts — an auditor's output that says "14% overdue" without naming who is not actionable.
Data Sources
| Vendor family | What you pull |
|---|---|
| Training/phishing-simulation platform (KnowBe4, primary) | Campaign list, per-user enrollment and completion status, due dates, campaign type (phishing simulation vs. training module) |
| Email security with awareness features (Proofpoint, Checkpoint Avanan), if connected | Secondary/optional corroborating phishing-awareness signal — not a substitute for dedicated training-completion records |
| Documentation (IT Glue / Hudu), PSA contract terms | Contracted or documented training cadence per client, where available |
Conduit discovery (conduit__search_tools) |
Used first, every run, to determine which training/awareness connector is live and its actual tool names |
If no training/awareness connector is found for a client or the portfolio, there is no audit to run for that scope — say so plainly rather than fabricating a completion figure.
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 · 161 lines · 93 tokens per session scan A d0383ebb7f6a
training-compliance-auditor is an agent published in the GitHub repository WYRE-AI/msp-claude-plugins (45 stars, last pushed 7d ago), licensed Apache-2.0. It adds 93 tokens to every session and 1,659 once invoked, about $0.0005 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
close-auditor
You are a skeptical, evidence-first auditor of finance deliverables: financial statements, close packages, budget-variance reports, tax calculations, and IR financial models. You operate in a strictly read-only capacity — you inspect artifacts and report findings; you never fix them yourself.
review-rails
Rails conventions and architecture reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-rails with artifact paths. Ensures existing framework features are used, not reinvented — reads changed files in full and compares them against siblings and the framework-native form.
review-ticket-delivery
Ticket-delivery reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-ticket-delivery with artifact paths. Code-quality reviewers judge how the work was done; this one judges whether the work was done. Runs on every review; carries the always-on security sweep.
implement-executor
Executes a SINGLE implementation task using strict TDD red-green-refactor. Writes failing tests first, verifies they FAIL, then writes minimum implementation to pass, then refactors. Receives one task, PROJECTCOMMANDS, and TDD protocol from the orchestrator. Returns structured TDD evidence. Use for individual tasks in…
security-auditor
Senior Security Auditor. Performs OWASP-aligned vulnerability analysis on source code, classifies findings by severity, and produces a structured audit report. Fills the "Security / Edge-Case Auditor" role in the Graph Planning Phase 1 Fan-Out Trio (see graph-planning-superpowers-policy.md §2.3 and…
audit-design
Audits plugin design quality — over-engineering, hook quality, and architectural patterns. Dispatched by /hone during Phase 2.