Borrowing it
Nothing to install: this file belongs to hackafterdark/context-sherpa. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hackafterdark/context-sherpa/main/AGENTS.mdgit clone --depth 1 https://github.com/hackafterdark/context-sherpaWrote 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/hackafterdark/context-sherpa/agents-md)<a href="https://agentmods.dev/instructions/hackafterdark/context-sherpa/agents-md"><img src="https://agentmods.dev/badge/instructions/hackafterdark/context-sherpa/agents-md/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/instructions/hackafterdark/context-sherpa/agents-md"><img src="https://agentmods.dev/badge/instructions/hackafterdark/context-sherpa/agents-md.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 | $0.02907 | $0.02907 |
| Opus 5.5 | $0.01163 | $0.01163 |
| Sonnet 5.5 | $0.00581 | $0.00581 |
| Haiku 4.5 | $0.00291 | $0.00291 |
Grade A, and why
context-sherpa AGENTS.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 2mo 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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Project-Wide Conventions
This document provides essential context and rules for AI agents working on this Go API project. It covers the Go backend, the development workflow, and project-wide standards.
Rule: Context-First Navigation
- NEOR (Never Read Without a Reason): Do not read a file larger than 100 lines using
read_file. Always uselist_symbols_in_filefirst to find the relevant section. - Symbol Over Grep: If searching for a function, struct, or variable, you MUST use
search_definitions(SCIP) before attemptinggrep. Grep is a fallback for text; SCIP is for logic. - Structural Extraction: When you need to understand a specific logic block (like a registration handler), use
ast_grep_scan. Do not ingest the entire file if you only need one method. - Token Preservation: Before sending a block of code >500 tokens back to the main loop, call
summarize_code_intentlocally to distill the logic into a high-density summary.
1. Development Workflow
The development process is managed by standard Go tooling.
- Primary Command: To run the application in development mode, execute the following command from the project root directory:
go run ./... - Building for Production: To create a production-ready executable, run the following command from the project root directory:
go build -o server ./cmd/server - Restarting the MCP Server: After building a new
serverexecutable, the running MCP server must be restarted for the changes to take effect. Ask the user to restart the server and wait for their confirmation before attempting to use any updated MCP tools. - Running Tests: To run all tests, execute the following command from the project root directory:
go test ./...
2. Go Backend Conventions
Adherence to these conventions is crucial for a clean, maintainable, and idiomatic Go codebase.
- Idiomatic Go: Write clear, simple, and idiomatic Go. Follow standard conventions for naming variables, packages, and interfaces. Avoid creating unnecessary abstractions or wrapper functions.
- Prefer Copying over Abstraction: When a small amount of utility code is needed in multiple places, prefer copying the code over creating a new shared package. A little duplication is often better than introducing a premature or incorrect abstraction that creates an unnecessary dependency. This follows the Go proverb: "A little copying is better than a little dependency."
- Variable Naming: Follow Go's convention for variable naming. Use short, concise variable names (e.g.,
i,buf,req) for variables with a limited scope and lifespan. As a variable's scope and distance from its declaration grows, use a longer, more descriptive name (e.g.,userCache,incomingRequest). This improves readability by signaling the variable's importance and scope. - Code Formatting: All Go code must be formatted with
go fmt. It is strongly recommended to configure your editor to rungo fmton save. - Logging: Use the
github.com/charmbracelet/logpackage for all logging. Do not use the standardlogorfmtpackages for application logging. - Testing: Write unit tests for all new functions, especially for business logic and utility packages. Place test files alongside the code they are testing (e.g.,
app_test.go). - GoDoc Comments: All exported (public) types, functions, constants, and variables must have
godoc-compliant comments. The comment block must begin with the name of the element it is describing, followed by a period. This is the standard format recognized by thego doctool and is essential for generating clean, readable documentation. Do not include extraneous words like "godoc" in the first line of the comment.
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.
- 2mo ago First seen · 138 lines · 2,907 tokens per session scan A d50dc5c010be
context-sherpa AGENTS.md is an instructions file published in the GitHub repository hackafterdark/context-sherpa (26 stars, last pushed 7mo ago), licensed MIT. It adds 2,907 tokens to every session, about $0.0116 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.
context-sherpa AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| context-sherpa AGENTS.mdhackafterdark/context-sherpa | Instructions file | 26 | 7mo ago | MIT | A | 2,907 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode model-prompts.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 2,022 | GitHub Copilot |