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/sovantica/engrava-mcp/copilot-instructionsgit clone --depth 1 https://github.com/sovantica/engrava-mcpWrote 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/sovantica/engrava-mcp/copilot-instructions)<a href="https://agentmods.dev/instructions/sovantica/engrava-mcp/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/sovantica/engrava-mcp/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 5.1 | $0.00539 | $0.00539 |
| Opus 5 | $0.00269 | $0.00269 |
| Sonnet 5 | $0.00108 | $0.00108 |
| Haiku 4.5 | $0.00054 | $0.00054 |
Grade A, and why
engrava-mcp 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 5d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Instructions — engrava-mcp
Public repository (engrava-mcp) — Model Context Protocol server exposing Engrava as agent memory.
License: MIT. Primary language: python 3.11+.
Commit messages
- Use Conventional Commits 1.0.0:
<type>(<scope>): <description>. SeeCLAUDE.mdfor the type table. - Use only durable public identifiers: this repo's release tags (
vX.Y.Z), this repo's GitHub issue/PR numbers (#N), and locally-documented ADRs (docs/adr/ADR-NNN-*.mdafter the file exists here). - Generic feature descriptions are welcome; concise imperative subject lines preferred (max 72 chars).
- Default PR target: the active
release/vX.Y.Zbranch. This repo has nodevbranch; the version is set by hand inpyproject.tomlandserver.json,CHANGELOG.mdis hand-maintained, and a release is a manualvX.Y.Ztag onmain.
Branch names
- Format:
<type>/<kebab-description>. Plain English. ≤ 50 chars. - No internal identifiers, no abbreviations that wouldn't make sense to someone reading the repo for the first time.
Code style
- Follow the existing patterns in
src/. Read 3-5 nearby files before introducing a new pattern. - Adhere to the quality bar described in
CLAUDE.md— linting clean, typing strict, coverage above threshold, no shortcuts. - Match the project's existing naming, file layout, and import conventions; don't invent new structures unprompted.
What to avoid
- Don't reference internal-looking identifiers or paths in commits, code comments, or PR descriptions. If something feels like it belongs to a private workflow, leave it out and the maintainer can add a public mapping (a
#Nissue/PR reference) later. - Don't introduce dependencies without checking the project's existing stack — extra deps need explicit justification.
- Don't suggest
--no-verifyto skip the local pre-commit hook except in maintainer-approved emergencies.
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.
- 5d ago First seen · 35 lines · 539 tokens per session scan A 572a8c287ca5
engrava-mcp copilot-instructions.md is an instructions file published in the GitHub repository sovantica/engrava-mcp (1 stars, last pushed 5d ago), licensed MIT. It adds 539 tokens to every session, about $0.0027 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 2026-08-31.
Other instructions, from other repositories
xbrain CLAUDE.md
Claude Code instructions for mrboups/xbrain, covering claude.md, project status, workflow: gsd is the build system, constraints carried over from kickoff and commands.
mistral-mcp CLAUDE.md
Instructions for Swih/mistral-mcp, covering claude.md — mistral-mcp, 1. identité du projet, 2. règles dures (ne pas transgresser), 3. layout & responsabilités and 4. conventions de code.
pfsense-mcp-server AGENTS.md
AGENTS.md instructions for night4me/pfsense-mcp-server, covering pfsense mcp server, architecture and security, development workflow, long-running validation and test parallelism.
backlog-mcp AGENTS.md
AGENTS.md instructions for gkoreli/backlog-mcp, covering agents.md — guidelines for ai agents working in backlog-mcp, testing, philosophy, how it works and test locations.
titen AGENTS.md
Instructions for RamaAditya49/titen, covering agents.md, product, repository stage, required work lifecycle and coding constraints.
swiss-culture-mcp CLAUDE.md
Claude Code instructions for malkreide/swiss-culture-mcp, covering claude.md, vor der arbeit, tests, wenn etwas rot ist and wenn codex gar nicht erst hinsieht.