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 instructions/acedatacloud/facetransformmcp/copilot-instructionsgit clone --depth 1 https://github.com/AceDataCloud/FaceTransformMCPWrote 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/instructions/acedatacloud/facetransformmcp/copilot-instructions)<a href="https://agentmods.dev/instructions/acedatacloud/facetransformmcp/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/acedatacloud/facetransformmcp/copilot-instructions.svg" alt="Measured on agentmods" 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 | $0.00413 | $0.00413 |
| Opus 5 | $0.00206 | $0.00206 |
| Sonnet 5 | $0.00083 | $0.00083 |
| Haiku 4.5 | $0.00041 | $0.00041 |
Grade A, and why
FaceTransformMCP copilot-instructions.md 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 3d 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.
This is a copy
88% identical to SunoMCP copilot-instructions.md — 10 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Copilot Instructions for FaceTransformMCP
Sync from Docs
When working on a sync task (issue labeled auto-sync), follow these rules:
- Source of truth — Clone/checkout
AceDataCloud/Docsand readopenapi/face.json(the OpenAPI spec). - Compare models — The Literal types in
core/types.pymust exactly match the model/enum values from the OpenAPI spec. Add missing values, remove deprecated ones. - Compare endpoints — Each endpoint in the spec should have a corresponding
@mcp.tool()function intools/. Add tools for new endpoints if needed. - Compare parameters — Each tool function's parameters should match the corresponding OpenAPI endpoint's request body schema.
- Update defaults — If a new model becomes the recommended default, update the default in
core/types.py. - Update README — Keep the model table and feature list in
README.mdcurrent. - Add tests — For new tools or parameters, add test cases in
tests/. - PR title — Use format:
sync: <description> [auto-sync] - No changes needed — If everything is already in sync, close the issue with a comment.
Project Structure
core/
config.py — Settings dataclass (API token, base URL)
server.py — FastMCP server singleton
client.py — httpx async HTTP client
types.py — Literal types (models, enums)
exceptions.py — Error classes
utils.py — Formatting helpers
tools/ — @mcp.tool() decorated async functions
prompts/ — LLM guidance prompts
tests/ — pytest-asyncio + respx tests
Development
pip install -e ".[dev]"
pytest --cov=core --cov=tools
ruff check .
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.
- 3d ago First seen · 39 lines · 413 tokens per session scan A 780aa8157c20
FaceTransformMCP copilot-instructions.md is an instructions file published in the GitHub repository AceDataCloud/FaceTransformMCP (0 stars, last pushed 11d ago), licensed MIT. It adds 413 tokens to every session, about $0.0021 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to SunoMCP copilot-instructions.md, differing in 10 lines, and is treated as a copy.
Other instructions, from other repositories
ditherer AGENTS.md
Instructions for gyng/ditherer, covering ditherer — agent guidelines, project overview, architecture, component hierarchy (atomic design) and state management.
obsidian-transcriber AGENTS.md
Instructions for dsebastien/obsidian-transcriber, covering project documentation, documentation surfaces, project overview, agent workflow and session-start checklist.
CShade AGENTS.md
Instructions for papadanku/CShade, covering cshade agent instructions, repo type, structure, file types and coding conventions (from readme.md).
ditherer CLAUDE.md
Instructions for gyng/ditherer, covering ditherer and quick reference.
darkly AGENTS.md
AGENTS.md instructions for darkly-art/darkly, a project described as: Forbidden Editor for Artists 😈.
imagesorcery-mcp GEMINI.md
Instructions for sunriseapps/imagesorcery-mcp: On each new session, please re-read the README.md file to ensure you have the most up-to-date project context and instructions.