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.
npx skills add wardawgmalvicious/agent-config --skill fabric-semantic-model-auditgit clone --depth 1 https://github.com/wardawgmalvicious/agent-configWrote 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/skills/wardawgmalvicious/agent-config/fabric-semantic-model-audit)<a href="https://agentmods.dev/skills/wardawgmalvicious/agent-config/fabric-semantic-model-audit"><img src="https://agentmods.dev/badge/skills/wardawgmalvicious/agent-config/fabric-semantic-model-audit/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/skills/wardawgmalvicious/agent-config/fabric-semantic-model-audit"><img src="https://agentmods.dev/badge/skills/wardawgmalvicious/agent-config/fabric-semantic-model-audit.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 | $0.00206 | $0.04223 |
| Opus 5 | $0.00103 | $0.02112 |
| Sonnet 5 | $0.00041 | $0.00845 |
| Haiku 4.5 | $0.00021 | $0.00422 |
Grade A, and why
fabric-semantic-model-audit scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
(an internal local-cli template), prefer it over raw curl for the second How it starts
The opening of the file, as written. The whole thing — 336 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing a semantic model
A review procedure over a finished model — stand back from it and
report on what it is, not on the line you are typing. Everything else in
the payload that touches a semantic model is authoring guidance:
fabric-tmdl and coding-tmdl.md own conventions, fabric-tmdl-api
owns deployment, code-review reviews a diff. This owns the corpus.
The output is a set of findings, not a rewrite. Do not edit the model unless asked separately.
1. Establish storage mode before anything else
This gates every memory claim you are about to make, and getting it wrong
is this skill's main way of producing a confidently wrong answer. Read
mode: on each table's partition in definition/tables/*.tmdl:
grep -rh "mode:" definition/tables/*.tmdl | sort | uniq -c
import, directLake, directQuery or dual. Mixed modes mean a
composite model, which in turn means source groups — and cross
source group relationships are limited (§5). Direct Lake has two
forms, on OneLake and on SQL analytics endpoint, and they differ
on nearly every limitation that matters to remediation; the connection
expression in expressions.tmdl tells you which.
State the storage mode in the finding report header. An audit that prescribes an import-mode fix to a Direct Lake model is wrong even when every individual sentence in it is true.
2. Three tiers of evidence
Not a fallback chain. Each tier answers questions the others cannot, and you should say which tier a finding came from.
| Tier | Needs | What it gets you |
|---|---|---|
TMDL on disk — definition/** |
nothing | Shape, relationships, declared cardinality, measure text, storage mode. Works offline and in CI. Most of §4 and §5 come from here. |
Live metadata — INFO.VIEW.* via executeQueries |
model admin + Build; tenant setting | The model after binding: real storage mode, calculated columns, measure [State]. |
Fabric notebook — sempy.fabric |
Fabric capacity, workspace contributor, ReadWrite on the model | Best Practice Analyzer (60+ rules, five categories) and Model Memory Analyzer. The only path to column-level memory and cardinality for a Direct Lake model. |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago Changed 14918b1e0675
- 7d ago Changed d08b453e5921
- 8d ago First seen · 336 lines · 206 tokens per session scan A 9ae7c7cd137c
fabric-semantic-model-audit is a skill published in the GitHub repository wardawgmalvicious/agent-config (1 stars, last pushed yesterday), licensed MIT. It adds 206 tokens to every session and 4,223 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
drawio
WORKFLOW SKILL — Generate Azure architecture diagrams in .drawio via simonkurtz-MSFT MCP server (full Azure icon set, batch creation, transactional mode). Covers architecture, dependency, runtime-flow, and as-built diagrams. WHEN: 'draw.io diagram', 'Azure architecture diagram', 'as-built diagram', 'runtime flow…
azure-artifacts
UTILITY SKILL — Artifact template structures, H2 compliance rules, and documentation styling for agent outputs (Steps 1-7). WHEN: "generate artifact", "check H2 structure", "artifact template", "step 7 as-built". USE FOR: generating any agent artifact, checking H2 structure compliance. DO NOT USE FOR: Azure resource…
azure-bicep-patterns
UTILITY SKILL — Reusable Azure Bicep patterns: hub-spoke, private endpoints, diagnostics, AVM composition. WHEN: "hub-spoke Bicep", "private endpoint module", "diagnostic settings", "AVM Bicep composition". USE FOR: Bicep template design, hub-spoke networking, private endpoint patterns, AVM modules. DO NOT USE FOR…
azure-governance-discovery
ANALYSIS SKILL — Azure Policy discovery: effective assignments (incl. MG-inherited), definitions/exemptions, effect classification, emits governance-constraints JSON. WHEN: 'Azure policy discovery', 'effective policy assignments', 'governance constraints', '04g-Governance Phase 1', 'refresh governance JSON'. DO NOT…
vendor-prompting
ANALYSIS SKILL — Audit-grade reference for Anthropic Claude and OpenAI GPT-5.6 prompting best practices. WHEN: "claude prompting", "gpt-5.6 prompting", "audit agent", "review prompt", "vendor best practices", "anthropic best practices", "openai prompting". DO NOT USE FOR: routine prompt edits where rules are already…
copilot-customization
Authoritative reference for VS Code Copilot customization mechanisms: instructions, prompt files, custom agents, agent skills, MCP servers, hooks, and plugins. Use when deciding which customization type to use, creating new .instructions.md/.prompt.md/.agent.md/SKILL.md/mcp.json files from scratch, or debugging why a…