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/cyberash-dev/veche/agents-mdgit clone --depth 1 https://github.com/cyberash-dev/vecheWhat 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 | $0.04387 | $0.04387 |
| Opus 5 | $0.02194 | $0.02194 |
| Sonnet 5 | $0.00877 | $0.00877 |
| Haiku 4.5 | $0.00439 | $0.00439 |
Grade B, and why
veche AGENTS.md scanned grade B with 2 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
(`~/.claude.json`, `~/.codex/config.toml`, `~/.hermes/config.yaml`) are NEVER edited Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
`node:child_process` for `--open`). If a dep feels needed, ask first. How it starts
The opening of the file, as written. The whole thing — 307 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Conventions for any AI agent (Codex, Claude Code, Cursor, Aider, …) editing this repository.
This file is the single source of truth; CLAUDE.md points here.
The non-negotiable rule
Spec-Driven Development. The specification in spec/spec.md is the contract.
Code that diverges from the spec is a bug.
The workflow for any non-trivial change is:
- Read the relevant spec file(s).
- Update the spec first. Get human approval if the change is substantive.
- Only then modify code to match.
- Run
npm run typecheck && npm test && npm run lintbefore declaring done.
Trivial changes (typo, obvious one-line fix, dependency bump) can skip step 2. When in doubt, change the spec first.
Project layout
veche/
├── spec/
│ └── spec.md ← single monolithic contract; sections 1–19 +
│ `## Partition: <name>` per feature
│ (persistence, agent-integration,
│ committee-protocol, meeting, web-viewer, install)
├── src/ ← implementation, hexagonal + vertical slices
│ ├── features/<slice>/ ← one folder per Partition
│ ├── adapters/inbound/mcp/ ← MCP server (stdio)
│ ├── adapters/inbound/cli/ ← human-operator CLI (list, show, watch, install)
│ │ + renderers + lib helpers
│ ├── adapters/inbound/web/ ← `veche watch` HTTP/SSE server
│ ├── infra/ ← composition root (DI wiring)
│ ├── bin/veche-server.ts ← stdio MCP entrypoint
│ └── bin/veche.ts ← CLI entrypoint
├── skills/veche/ ← canonical SKILL.md + optional UI metadata
├── examples/ ← sample .mcp.json, config.json
├── scripts/ ← repo helpers (e.g. sdd-approve-all.sh)
├── .sdd/ ← SDD config + finalised approval plans
├── dist/ ← build output (not committed locally)
└── node_modules/
Navigation: open spec/partitions/<name>.md for the slice you're touching.
Each partition file is self-contained (Context, Glossary, Surfaces,
Behaviours, Contracts, Invariants, Policies, Constraints, Migrations,
Deltas, Open questions, Assumptions, Out-of-scope). The thin
spec/spec.md is just an index.
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 · 307 lines · 4,387 tokens per session scan B 447a89f63d2c
veche AGENTS.md is an instructions file published in the GitHub repository cyberash-dev/veche (21 stars, last pushed 3mo ago), licensed MIT. It adds 4,387 tokens to every session, about $0.0219 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
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
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, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain 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.