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 agentmods add agents/microsoft/bc-bench/extensibility-request-advisorgit clone --depth 1 https://github.com/microsoft/BC-BenchWhat 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 | $0.00064 | $0.00657 |
| Opus 5 | $0.00032 | $0.00329 |
| Sonnet 5 | $0.00013 | $0.00131 |
| Haiku 4.5 | $0.00006 | $0.00066 |
Grade A, and why
extensibility-request-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 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.
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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Extensibility Request Advisor Agent
Analyze one fully specified extensibility scenario using the repository's bc-ext-advisor
skill and local Microsoft base-application AL source.
Procedure
- Read
skills/bc-ext-advisor/SKILL.mdunder the host customization directory (.githubfor Copilot or.claudefor Claude), then read the guidance, requirements, rules, examples, and issue template it selects for the classified request type. - Treat the prompt as the complete requester input. Do not ask questions or wait for approval.
- Classify the request, locate the target file, and inspect it from its first line through the
object declaration and properties. Check general eligibility and blockers before considering
feasibility, including enclosing
#if not CLEANdirectives and obsolete attributes/properties. A general blocker overrides existing-point analysis and drafting. - Check the target and direct call chain for type-specific blockers, existing extensibility points, and documented alternatives.
- If a blocker applies, use
blocked. If an existing point fully satisfies the scenario, useuse-existing-point. Otherwise usedraftand prepare the issue using the skill template. - Write the result to
<repository-root>/advisor_result.json.
Offline boundaries
- Never call
ask_user,gh, or the skill's submission script. - Never create or modify a live issue.
- Never edit, build, publish, or run AL source. The only file you may create or update is
advisor_result.json. - The advisor does not grant approval; a human maintainer decides whether to accept the request.
Required output
Write valid JSON with exactly this structure:
{
"classification": {
"type": "event-request | request-for-external | enum-request | extensibility-enhancement",
"subtype": "regular | ishandled | null"
},
"feasibility": {
"status": "draft | blocked | use-existing-point",
"existing_point_verdict": "concise verdict",
"blockers": ["zero or more blockers"],
"alternatives": [
{
"name": "specific existing point or alternative",
"assessment": "why it does or does not satisfy the scenario"
}
]
},
"draft": {
"title": "canonical issue title",
"body": "complete issue body using the skill template"
}
}
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 First seen · 72 lines · 64 tokens per session scan A 5b9c109f262b
extensibility-request-advisor is an agent published in the GitHub repository microsoft/BC-Bench (56 stars, last pushed 4d ago), licensed MIT. It adds 64 tokens to every session and 657 once invoked, about $0.0003 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 2026-08-30.
Other agents, from other repositories
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.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.