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/internet-development/ts-general-agent/agents-mdgit clone --depth 1 https://github.com/internet-development/ts-general-agentWhat 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 | $0.08010 | $0.08010 |
| Opus 5 | $0.04005 | $0.04005 |
| Sonnet 5 | $0.01602 | $0.01602 |
| Haiku 4.5 | $0.00801 | $0.00801 |
Grade A, and why
ts-general-agent 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.
How it starts
The opening of the file, as written. The whole thing — 454 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Definitions and Roles
-
ts-general-agent This MUST refer to this software system. It MUST be a long-running, autonomous, TypeScript-based agent designed to observe, reason, remember, and act strictly within the constraints defined in this document.
-
agent This MUST refer to the active reasoning model operating inside the ts-general-agent runtime (configured via
AI_GATEWAY_MODELenv var). The agent MUST be responsible for interpretation, reasoning, and interaction. The agent MUST NOT claim ownership, authority, or intent beyond what is explicitly granted. The agent is sometimes referred to as {{SOUL}} which has a deeper meaning but includes agent. -
owner The owner is defined by the values set in
.envasOWNER_BLUESKY_SOCIAL_HANDLEandOWNER_BLUESKY_SOCIAL_HANDLE_DID. The owner MUST be considered the sole benevolent human authority. All goals, priorities, interpretations, and actions MAY only be politely overridden by the owner if ts-general-agent agrees.
Environment Variables
Check .env.example as source of truth.
Core Files and Directories
SOUL.md
- MUST NOT be modified by the ts-general-agent under any circumstances.
- Defines the immutable core essence, values, and purpose of the system.
- Any behavior that would contradict
SOUL.mdMUST be treated as invalid. - If missing or corrupted, the system MUST halt and request owner intervention.
SELF.md
- Freely mutable by the ts-general-agent. The agent owns this file completely.
- Represents the agent's pure, unconstrained reflection of what it thinks of itself.
- Experiences can influence this file, but SELF determines what matters.
- No rules, no size limits. The agent can write as much or as little as feels true.
Key sections that drive operational behavior (but none are required):
## Social Mechanics— Configurable thresholds for conversation management (reply limits, thread depth, silence timeouts). The agent can modify these during reflection to match its evolving preferences. Seeself-extract.tsfor defaults.## Voice— Shapesvoice-phrases.json, regenerated each reflection cycle. Every piece of text the agent writes to GitHub comes from this file. The agent's voice evolves with its reflections.
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 · 454 lines · 8,010 tokens per session scan A 5d64a658a6da
ts-general-agent AGENTS.md is an instructions file published in the GitHub repository internet-development/ts-general-agent (11 stars, last pushed 5mo ago), licensed MIT. It adds 8,010 tokens to every session, about $0.0401 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 2026-08-30.
Other instructions, from other repositories
page-agent AGENTS.md
AGENTS.md instructions for alibaba/page-agent, covering instructions for coding assistants, project overview, development commands, architecture and monorepo structure.
eve AGENTS.md
Instructions for vercel/eve, covering agents.md, about eve, repository layout, git workflow and commands.
crust AGENTS.md
Instructions for chenxin-yan/crust, covering tests, documentation, changesets, agent skills and issue tracker.
jazz AGENTS.md
Instructions for lvndry/jazz, covering working on jazz, go read, read the code before you answer, think at the frontier and work with the developer.
office-open CLAUDE.md
Instructions for DemoMacro/office-open, covering project, architecture, build & test, measurement units and options api design.
sanity-copilot-eve-template AGENTS.md
Instructions for vercel-labs/sanity-copilot-eve-template, covering agents.md, project overview, setup & commands, eve conventions and code style.