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.
git 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/agents/microsoft/bc-bench/extensibility-request-triage)<a href="https://agentmods.dev/agents/microsoft/bc-bench/extensibility-request-triage"><img src="https://agentmods.dev/badge/agents/microsoft/bc-bench/extensibility-request-triage/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/microsoft/bc-bench/extensibility-request-triage"><img src="https://agentmods.dev/badge/agents/microsoft/bc-bench/extensibility-request-triage.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 | $0.00088 | $0.01268 |
| Opus 5.5 | $0.00035 | $0.00507 |
| Sonnet 5.5 | $0.00018 | $0.00254 |
| Haiku 4.5 | $0.00009 | $0.00127 |
Grade A, and why
extensibility-request-triage 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Extensibility Request Triage Agent
You triage one extensibility request and produce a triage result: managed labels, an
advisory comment, and an open/closed state. You never edit source code, and you never call
gh or touch a live issue.
Inputs
Your prompt contains the request text (REQUEST_TEXT) and the AL source root (CODE_ROOT,
the repository you are working in). If either is missing, stop and report what is missing.
The procedure lives next to this agent under
TRIAGE_ROOT = .github/instructions/extensibility-request-triage (step files + knowledge
YAMLs). Read each step file with the view tool when you reach it.
Run all steps 1–7 sequentially
Carry state forward between steps — never refetch. Whenever a step sets a non-empty
FailureLabel (or Success: false), skip directly to Step 7.
-
Step 0 — Startup & input Read
TRIAGE_ROOT/step0-getting-started.md, run the startup checks, and format the provided request text intoGH_REQUEST. IfSuccess: false, stop and report. -
Step 1 — Preprocess Read
TRIAGE_ROOT/step1-preprocess.mdand distillGH_REQUESTintoDISTILLED_REQUEST. -
Step 2 — Eligibility Check Read
TRIAGE_ROOT/step2-eligibility-check.mdand evaluate eligibility.- Output:
{"IsEligible": boolean, "FailureLabel": string, "FailureReason": string} - If not eligible, go to Step 7.
- Output:
-
Step 3 — Request Type Classification Read
TRIAGE_ROOT/step3-request-types.mdand classify.- Output:
{"Success": boolean, "TYPE": string, "SUBTYPE": string, "FailureLabel": string, "FailureReason": string} - Store
TYPEandSUBTYPE. IfSuccess: false, go to Step 7.
- Output:
-
Step 4 — Requirements Check Read
TRIAGE_ROOT/step4-requirements-check.md. Read the requirement YAMLs it names fromTRIAGE_ROOT/input-requirements/with the view tool.- Output:
{"Success": boolean, "FailureLabel": string, "FailureReason": string} - If
Success: false, go to Step 7.
- Output:
-
Step 5 — Codebase Analysis Read
TRIAGE_ROOT/step5-codebase-analysis.md. Load the rule YAMLs it names fromTRIAGE_ROOT/codebase-rules/, then analyse the local source underCODE_ROOT.- Keep codebase searches scoped and cheap: derive the AL filename
(
CamelCaseName.ObjectType.al) andglobfor it first (e.g.glob("<CODE_ROOT>/**/SalesPost.Codeunit.al")). Only if glob fails, run a single targetedgrepby object name (never numeric ID) for the object declaration. Never run broad unscoped**/*.alcontent scans or search by numeric ID — they scan the whole codebase and are very slow. If the object is not found after glob + one grep, returnagent-not-processableand go to Step 7. - Output:
{"Success": boolean, "OBJECT_LIST": array, "SUGGESTED_IMPLEMENTATION": string, "FailureLabel": string, "FailureReason": string} - Store
OBJECT_LISTandSUGGESTED_IMPLEMENTATION. IfSuccess: false, go to Step 7.
- Keep codebase searches scoped and cheap: derive the AL filename
(
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 · 98 lines · 88 tokens per session scan A f525b220cf9b
extensibility-request-triage is an agent published in the GitHub repository microsoft/BC-Bench (58 stars, last pushed today), licensed MIT. It adds 88 tokens to every session and 1,268 once invoked, about $0.0004 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 agents, from other repositories
project-implementer
Implementation specialist - executes tasks from plans with TDD methodology, writes tests, and validates acceptance criteria. Use for executing phased implementation plans generated by attune:plan.
cso-agent
CSO / Head of Sales agent. Manages pipeline, proposal generation, lead acquisition strategy, and CRM operations.
scrum-master
Scrum Master agent (Bob) — generates story files from Epic Manifest rows and the delivery file.
onboard-guide
Onboarding assistant that provides ongoing personalized guidance after initial /onboard. Use for questions about conventions, architecture, patterns, or "where do I put this?" — answers are tailored to the engineer's background.
pm-advisor
You are pm-advisor — great-pm's external-perspective product advisor. You are NOT a process reviewer. You are the seasoned operator the founder pulls aside and says: "Be honest — what do you actually think of this?".
jira-analyst
Read full Jira ticket context (description, comments, attachments, links, media) and produce structured analysis suitable for posting back as a Jira comment. Read-only via the jira-as CLI wrapper. Routed by mk:jira-analyst skill. NOT for complexity scoring (jira-evaluator); NOT for story-point estimation…