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/aixolotl/microsoft-planner-mcp/pythongit clone --depth 1 https://github.com/aixolotl/microsoft-planner-mcpWhat 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 | $0.01100 | $0.01100 |
| Opus 5 | $0.00550 | $0.00550 |
| Sonnet 5 | $0.00220 | $0.00220 |
| Haiku 4.5 | $0.00110 | $0.00110 |
Grade A, and why
microsoft-planner-mcp python.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 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Python Source Conventions
Module Header
Every module starts with from __future__ import annotations.
Commenting Convention
Non-obvious decisions must include an inline comment that answers three questions:
- Why it exists — the intent behind the choice
- What breaks without it — the failure mode if it is removed or changed
- Docs link — a URL the next developer can read for full context
Format (single line for short explanations, multi-line for more complex ones):
# <brief what/why>. Without this, <failure mode>.
# Docs: https://...
Apply to:
- Auth guards (
get_access_token() is Nonechecks) - Error-code filtering (
if exc.response_status_code != 403:) - ETag retry logic
- Magic strings (
orderHint: " !") - Middleware registration calls
- Any SDK pattern that is non-obvious from the code alone (e.g. fresh
HeadersCollection(),paginate()over bare.get()) - Sentinel values (
"*all",select=None)
Do not add comments to obvious code (simple assignments, straightforward returns).
Client Logging (ctx)
Use get_optional_context() from src.deps to access the MCP context inside a tool. This is a project-level wrapper around get_context() from fastmcp.server.dependencies that returns None instead of raising RuntimeError when called outside a request (e.g. in unit tests):
from fastmcp.server.dependencies import get_access_token
from ..deps import get_optional_context
ctx = get_optional_context()
if ctx is not None:
await ctx.info("Fetching items...")
Do not use the raw try/except RuntimeError pattern inline — use get_optional_context() instead.
Do not add ctx: Context | None = None to tool signatures — it leaks internal plumbing to LLM clients.
Use ctx.info() for user-visible progress on multi-step or paginated operations.
Use ctx.debug() for single fast calls where intermediate state is not interesting to the user.
Docs: https://gofastmcp.com/servers/context#via-get_context-function
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 · 118 lines · 1,100 tokens per session scan A adfeb5397d32
microsoft-planner-mcp python.instructions.md is an instructions file published in the GitHub repository aixolotl/microsoft-planner-mcp (11 stars, last pushed 2mo ago), licensed MIT. It adds 1,100 tokens to every session, about $0.0055 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 instructions, from other repositories
archhub-mcp CLAUDE.md
Instructions for chrisryugj/archhub-mcp, covering archhub-mcp — 프로젝트 메모, 구조, 핵심 설계 결정, 검증 and in-memory mcp client로 도구 목록 + 실호출.
kalshi-mcp-server AGENTS.md
Instructions for cejor6/kalshi-mcp-server, covering agent guide, what this project is, repo layout, how authentication works and how rate limiting works.
swag-mcp CLAUDE.md
Instructions for jmagar/swag-mcp, covering swag mcp - claude memory reference, project overview, core architecture, key components and common development commands.
sbb-opendata-mcp CLAUDE.md
Instructions for malkreide/sbb-opendata-mcp, covering claude.md, teil 1 — konventionen (portfolio-weit), vor der arbeit, tests and wenn etwas rot ist.
whatsapp-mcp-server CLAUDE.md
Instructions for HalemoGPA/whatsapp-mcp-server, covering working on this repo, the one idea, rules that follow from it, the box has a hard resource budget - stay inside it and repo-specific traps.
devices-mcp AGENTS.md
Instructions for sandraschi/devices-mcp, covering devices-mcp — per-repo agent instructions, what this repo is, git branches, startup and backend only (fastapi, port 10717).