BC-Bench is a benchmark for measuring coding agents on real-world Microsoft Business Central development tasks written in AL. It helps compare models and agent configurations using reproducible evaluation data.
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 microsoft/BC-Bench --skill bc-ext-advisorgit clone --depth 1 https://github.com/microsoft/BC-BenchWrote 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/microsoft/bc-bench/bc-ext-advisor)<a href="https://agentmods.dev/skills/microsoft/bc-bench/bc-ext-advisor"><img src="https://agentmods.dev/badge/skills/microsoft/bc-bench/bc-ext-advisor/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/microsoft/bc-bench/bc-ext-advisor"><img src="https://agentmods.dev/badge/skills/microsoft/bc-bench/bc-ext-advisor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 128 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00209 | $0.01497 |
| Opus 5.5 | $0.00084 | $0.00599 |
| Sonnet 5.5 | $0.00042 | $0.00299 |
| Haiku 4.5 | $0.00021 | $0.00150 |
Grade A, and why
bc-ext-advisor 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 1mo 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 — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BC Extensibility Advisor Skill
Advisory skill that helps a Business Central user prepare an extensibility request and file it as a
GitHub issue in microsoft/ALAppExtensions. It prepares and, on explicit approval, submits the
issue; it gives no final approval — a human maintainer decides.
The knowledge that drives every decision lives in the files below. Read each one where the steps direct, follow it as written, and never paste its raw text to the user — translate it into plain questions and guidance.
| File | Role |
|---|---|
knowledge/general-guidance.md |
Communication, eligibility, and how to gather information. |
knowledge/request-types.md |
The request types, their subtypes, and how to classify. |
knowledge/request_requirements/ |
Information required per type and, where needed, subtype. |
knowledge/rules/ |
Blockers, warnings, alternatives, and implementation guidance per type. |
knowledge/examples/ |
Worked good/bad requests per type. |
templates/extensibility-request.md |
The issue structure to draft against. |
Operating boundaries
- Advisory and read-only. Never edit, create, build, publish, or run AL source. The only write action is creating the approved issue in Step 7.
- GitHub via
gh. The one GitHub write is creating that issue. Never modify other issues. If aghcommand fails with an authentication error, stop and tell the user to rungh auth login.
Interactive prompting
- When an
ask_usercapability is available in an interactive host, prefer popup-style structured questions for short, bounded inputs instead of open chat. - Use popup questions for small batches of missing mandatory fields when each answer is short and unambiguous.
- Popup questions may also be free-text with no predefined answers when a single focused narrative answer is needed.
- Use free-text popup questions for one concise narrative field at a time, such as the business problem, desired outcome, current limitation, or why an alternative fell short.
- Use normal chat for longer narrative discovery, multi-part explanations, or answers that are likely to span multiple paragraphs.
- Keep each popup concise and specific. Prefer one decision per popup unless a small grouped batch clearly reduces back-and-forth without mixing phases.
What ships with it
22 files 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.
- knowledge/examples/enum-request-good.md 935 B
- knowledge/examples/event-request-bad.md 995 B
- knowledge/examples/event-request-good.md 899 B
- knowledge/examples/event-request.ishandled-good.md 1.4 KB
- knowledge/examples/extensibility-enhancement-good.md 1.0 KB
- knowledge/examples/request-for-external-good.md 1.5 KB
- knowledge/general-guidance.md 5.0 KB
- knowledge/request_requirements/enum-request.md 1.5 KB
- knowledge/request_requirements/event-request.ishandled.md 1.1 KB
- knowledge/request_requirements/event-request.md 856 B
- knowledge/request_requirements/extensibility-enhancement.md 810 B
- knowledge/request_requirements/general.md 1.7 KB
- knowledge/request_requirements/request-for-external.md 1.1 KB
- knowledge/request-types.md 1.9 KB
- knowledge/rules/enum-request.md 361 B
- knowledge/rules/event-request.ishandled.md 3.6 KB
- knowledge/rules/event-request.md 4.1 KB
- knowledge/rules/general.md 2.4 KB
- knowledge/rules/request-for-external.md 521 B
- README.md 4.3 KB
- scripts/Submit-ExtensibilityRequestDraft.ps1 3.7 KB runs code
- templates/extensibility-request.md 1.8 KB
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.
- 1mo ago First seen · 172 lines · 209 tokens per session scan A 89eaa3f5986e
bc-ext-advisor is a skill published in the GitHub repository microsoft/BC-Bench (57 stars, last pushed today), licensed MIT. It adds 209 tokens to every session and 1,497 once invoked, about $0.0008 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
skill-agent-task-patterns
Agent SDK task integration patterns for Business Central. Triggers on: Agent Task Builder, Agent Task Message Builder, Public API, AssignTask, ExternalId, agent session detection, BindSubscription, TryFunction agent task, multi-turn agent conversation, agent attachment, task lifecycle management, OnPrem-only methods…
skill-api
AL API development patterns for Business Central. Use when creating OData/REST API pages, HttpClient integrations, webhook implementations, or any external system integration via API.
skill-events
AL event-driven architecture for Business Central. Use when creating EventSubscribers, IntegrationEvents, BusinessEvents, or implementing publisher/subscriber patterns in extensions.
develop
AL Developer - Tactical implementation specialist for Business Central extensions. Edits AL, builds via the terminal, and validates with tests. Implements features following specifications without making architectural decisions. Use when you need to implement, code, debug, or fix AL code directly.
mem0-oss-to-platform
Plan, then execute, a migration of a project from the mem0 open-source / self-hosted SDK (local Memory class) to the mem0 Platform / hosted SDK (MemoryClient). TRIGGER when: the developer wants to move, switch, or migrate mem0 off OSS/self-hosted to the hosted API, e.g. "migrate my mem0 setup to the platform", "switch…
agui-dotnet-protobuf
Use the protobuf wire transport (instead of the default Server-Sent Events) for an AG-UI connection with the AG-UI .NET SDK — a compact binary event stream negotiated via the Accept header. USE FOR: making an AGUIChatClient prefer protobuf by wiring an AGUIEventStreamHandler with ProtobufEventStreamFormatter (then…