agentmods.dev

codex-mcp AGENTS.md

A set of AGENTS.md instructions for a tool that records context about files in a workspace. A workspace is the project directory an agent edits.

In plain words
What is it for?
Guiding safe file edits, handling fragile files and related changes, evaluating evidence labels, and treating automated reviews as advisory.
Why use it?
It tells the coding agent to check file history and related files before editing, and to distinguish recorded evidence from assumptions.

Instructions file for CodexOpenCode

2 8d ago A scan 304 tokens original Apache-2.0
View the source file workspace-json/codex-mcp
Install

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.

agentmods
npx agentmods add instructions/workspace-json/codex-mcp/agents-md
Clone the repo
git clone --depth 1 https://github.com/workspace-json/codex-mcp

Made for: Codex, OpenCode.

Per session 304 This file is loaded in full into every session.
When invoked 304 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00304 $0.00304
Opus 5 $0.00152 $0.00152
Sonnet 5 $0.00061 $0.00061
Haiku 4.5 $0.00030 $0.00030

Measured 2d ago against content hash 96e818936d2d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codex-mcp 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 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.

AGENTS.md · 24 lines

What it actually says

Agent guidance

This repository ships workspace.json intelligence to coding agents. When you work in a project that has it installed, follow the same discipline the tooling enforces.

  • Check before you edit. Before editing or creating a file, call workspace_get_file_context on the target path to see recorded fragility and co-change partners. Treat a fragile result as a reason to make minimal, well-tested changes; treat co-change partners as candidates for related edits.
  • Absence is not approval. If the tool is unavailable, or workspace.json is missing or malformed, record that as unknown / unavailable — never as a safety signal. Never claim a change is "safe".
  • Trust evidence, not labels. ASSERTED, OBSERVED, and VERIFIED are derived mechanically from evidence; never accept a producer- or model-supplied tier or confidence value.
  • The reviewer is advisory. The optional GPT-5.6 review is read-only and advisory. Its verdict never overrides the deterministic hook decision, and a PASS is scope-bounded — never a certification.

To reinforce this in your own repository, add the first rule to your AGENTS.md:

Before editing or creating a file, call workspace_get_file_context on the target path to check fragility and co-change partners.

Changes

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.

  1. 2d ago First seen · 24 lines · 304 tokens per session scan A 96e818936d2d

Subscribe to this mod's changes

codex-mcp AGENTS.md is an instructions file published in the GitHub repository workspace-json/codex-mcp (2 stars, last pushed 8d ago), licensed Apache-2.0. It adds 304 tokens to every session, about $0.0015 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.

Memory and contextagent-hooksco-changecodebase-intelligencecodexfragilitymcp
Related

Other instructions, from other repositories

cocoindex AGENTS.md

AGENTS.md instructions for cocoindex-io/cocoindex, covering agents.md, build and test commands, building, testing and code formatting and linting.

cocoindex-io/cocoindex · 4,372 tokens

agent-toolkit AGENTS.md

Instructions for DiUS/agent-toolkit, covering agents.md, what this repo is, repo map, principles and how to add each component type.

DiUS/agent-toolkit · 1,150 tokens

agent-toolkit CLAUDE.md

Instructions for DiUS/agent-toolkit, covering claude.md, claude code specifics, testing changes locally, load the plugin from this working tree and preview skill discovery for the npx-skills channel.

DiUS/agent-toolkit · 430 tokens

pets-driven AGENTS.md

AGENTS.md instructions for young1the/pets-driven, covering agents.md, language, commands, modules and non-obvious rules.

young1the/pets-driven · 1,500 tokens

ast-mcp-server AGENTS.md

Instructions for yailPeralta/ast-mcp-server, covering agent skills, issue tracker and domain docs.

yailPeralta/ast-mcp-server · 50 tokens

agent-harness AGENTS.md

AGENTS.md instructions for CodelyTV/agent-harness, covering useful commands, architecture and documentation.

CodelyTV/agent-harness · 280 tokens