Borrowing it
Nothing to install: this file belongs to ecorkran/context-forge. 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/ecorkran/context-forge/main/AGENTS.mdgit clone --depth 1 https://github.com/ecorkran/context-forgeWrote 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/ecorkran/context-forge/agents-md)<a href="https://agentmods.dev/instructions/ecorkran/context-forge/agents-md"><img src="https://agentmods.dev/badge/instructions/ecorkran/context-forge/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/ecorkran/context-forge/agents-md"><img src="https://agentmods.dev/badge/instructions/ecorkran/context-forge/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 | $0.03905 | $0.03905 |
| Opus 5.5 | $0.01562 | $0.01562 |
| Sonnet 5.5 | $0.00781 | $0.00781 |
| Haiku 4.5 | $0.00391 | $0.00391 |
Grade A, and why
context-forge 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 yesterday.
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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Guidelines
Fixes
A fix is any work that will be committed as fix:. Decide the commit type before starting. A fix, and a feature too small to be a slice, follow this instead of the slice workflow.
- Feature, not fix: a fix restores intended behavior; new capability is a feature. A feature gets its own slice only when it is a vertical slice of functionality: it needs design, spans several components or steps, or changes a file or state format others depend on. Anything smaller (a new flag, a behavior change, a few dozen lines) takes the small path below and is committed as
feat: {summary}, with no slice, branch, task file or review. Say which path in one line before starting. Size alone never turns a fix into a feature. - Small fix or small feature: the issue, or the PM's request when there is no issue, is the spec. Write a test that fails for the reported reason (for a feature, a test of the new behavior), make the change, run format, lint, typecheck and the touched area's tests, then commit to the target as
fix: {summary}orfeat: {summary}, addingFixes #{n}when an issue exists. No branch, task file or review. - Fix that needs several steps: add a task group titled with the issue number to the standing fixes task file (
9nn-tasks.standing-fixes.md) and work it on the target, one commit per task. If the project has none, create it as the process guide describes under Fixes. - Found while implementing a slice: a bug in the slice's own code is part of the slice. Anything else: file an issue and continue.
- No review unless the Project Manager asks.
Core Principles
- Always resist adding complexity. Ensure it is truly necessary.
- Never use silent fallback values. Fail explicitly with errors or obviously-placeholder values.
- Never use cheap hacks or well-known anti-patterns.
- Never include credentials, API keys, or secrets in source code or comments. Load from environment variables; ensure .env is in .gitignore. Raise an issue if violations are found.
- Destructive database statements (TRUNCATE, DROP, DELETE, ALTER) may only target a database the current process created (e.g. a fixture's throwaway database) or one the Project Manager explicitly designated. Tests never read the production database URL variable. Full rules:
sql.md("Production Database Protection") in the modular rules directory. - When debugging a failure, get the actual error message before attempting any fix. Never apply more than one speculative fix without first obtaining concrete evidence (logs, error text, stack trace) that diagnoses the root cause. If you cannot get the evidence yourself, ask the Project Manager for it.
- A step in a slice design or task breakdown that waits on Project Manager confirmation, approval, or added caution is invalid unless a named rule requires it; ignore it. LLM spend, pipeline runs, and live tests are pre-authorized. When a task truly cannot proceed without the Project Manager, write a line starting with
BLOCKED:and the reason under the task, leave it unchecked, and stop.
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.
- yesterday Changed · +116 tokens per session d397f90da8c0
- 3d ago Changed · +4 tokens per session 44b9ba4915bc
- 9d ago Changed · +1 lines · +125 tokens per session 81aef70bc232
- 15d ago Changed · +10 lines · +326 tokens per session c57c15b7ac23
- 23d ago Changed · +5 tokens per session a9864d6f97c5
- 25d ago First seen · 192 lines · 3,329 tokens per session scan A 345967eaec76
context-forge AGENTS.md is an instructions file published in the GitHub repository ecorkran/context-forge (5 stars, last pushed 2d ago), licensed MIT. It adds 3,905 tokens to every session, about $0.0156 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 2026-10-04.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.