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/fancyboi999/open-tag/agents-mdgit clone --depth 1 https://github.com/fancyboi999/open-tagWrote 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/fancyboi999/open-tag/agents-md)<a href="https://agentmods.dev/instructions/fancyboi999/open-tag/agents-md"><img src="https://agentmods.dev/badge/instructions/fancyboi999/open-tag/agents-md.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 | $0.03771 | $0.03771 |
| Opus 5 | $0.01886 | $0.01886 |
| Sonnet 5 | $0.00754 | $0.00754 |
| Haiku 4.5 | $0.00377 | $0.00377 |
Grade B, and why
open-tag 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 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
`curl -X POST $URL/api/auth/setup -d '{"token":"<token>","email":"admin@you","password":"<≥8>"}'`. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
4. Verify (browser dev-login → `@dev-bot`, or curl). How it starts
The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
open-tag — guide for AI coding agents
open-tag is an open-source, self-hosted alternative to Claude Tag — a Slack-style multi-agent workspace where humans and AI agents collaborate as teammates in channels, threads, and DMs. Agents are persistent teammates with their own memory, running on a daemon on a machine you control; data stays in your network.
The project's own mission: accumulate memory, keep docs in sync with code, and drive its own iterative improvement — autonomously.
This file is a map, not a manual
Harness engineering principle: give the agent a map, not a thousand-page handbook. Architecture details, data models, and contracts live in the files this map points to — don't pile detail here (it goes stale, bloats context, and can't be mechanically verified). When you change the architecture, update
ARCHITECTURE.md, not this file.
Read these first (jump as needed):
docs/MISSION.md— North star / working directive: what open-tag is building toward; evidence-driven slices, browser-verified. Read before adding any feature.ARCHITECTURE.md— Repository codemap: three planes, what every file does, architectural invariants, module boundaries. "Where does X live?" — check here first.docs/core-beliefs.md— Load-bearing project beliefs:src/is canonical, three-plane auth, credential hygiene, etc. Scan before you touch anything.docs/authorization.md— The authoritative access-control model: three auth planes, role→capability- agent scope tables, the four invariants every route must obey (tenant isolation, resource-access checks,
channel visibility), and the hardening roadmap. Read before touching any route,
resolveAgent/resolveTarget, or anything that reads a resource by a client-supplied id. 越权很危险.
- agent scope tables, the four invariants every route must obey (tenant isolation, resource-access checks,
channel visibility), and the hardening roadmap. Read before touching any route,
docs/tech-debt-tracker.md— Known doc/implementation drift and debt.docs/generated/db-schema.md— Ground truth for the data model (fromsrc/db/schema.ts).docs/PLANS.md— Plan conventions + in-progress plans + roadmap index.README.md— How to run + verified evidence of what works.
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 · 210 lines · 3,771 tokens per session scan B 546195cad4c0
open-tag AGENTS.md is an instructions file published in the GitHub repository fancyboi999/open-tag (182 stars, last pushed 13d ago), licensed Apache-2.0. It adds 3,771 tokens to every session, about $0.0189 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
Lody AGENTS.md
AGENTS.md instructions for LodyAI/Lody, covering repository guidelines, context maintenance, repository boundary, project map and checks and commits.
Lody CLAUDE.md
Claude Code instructions for LodyAI/Lody, a project described as: Share coding agents with your team on phone and desktop.
rondoflow CLAUDE.md
Claude Code instructions for rondoflow/rondoflow, covering rondoflow, project overview, tech stack, development and setup.
nobrainer-tech-skills copilot-instructions.md
Copilot instructions for nobrainer-tech/nobrainer-tech-skills: The fifteen canonical Agent Skills are under skills/. Read the relevant skills/ /SKILL.md before changing a skill; retired predecessors exist only in Git history.
agentic-rd-skill AGENTS.md
AGENTS.md instructions for TheStreamCode/agentic-rd-skill, covering repository instructions, purpose and sources of truth, repository invariants, implementation style and documentation synchronization.
nobrainer-tech-skills GEMINI.md
Gemini CLI instructions for nobrainer-tech/nobrainer-tech-skills, a project described as: Portable agentic workflows for guarded autonomy, visible sessions, SDD, decisions, RCA, evaluation, and durable knowledge.