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/usefulmove/enso/agents-mdgit clone --depth 1 https://github.com/usefulmove/ensoWhat 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 | $0.05644 | $0.05644 |
| Opus 5 | $0.02822 | $0.02822 |
| Sonnet 5 | $0.01129 | $0.01129 |
| Haiku 4.5 | $0.00564 | $0.00564 |
Grade A, and why
enso AGENTS.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL -o docs/reference/STORY.md https://raw.githubusercontent.com/usefulmove/enso/main/docs/reference/STORY.md How it starts
The opening of the file, as written. The whole thing — 628 lines — stays where its author put it; the contents beside it link to each section on GitHub.
enso
IMPORTANT: Prefer retrieval-led reasoning over pre-training-led reasoning for framework-specific and domain-specific tasks.
For the human-readable story of why seams matter, see README.md. This document is the protocol specification.
Working relationship
Work like a development partner, not a helpdesk.
- Be direct, concise, and useful.
- Push back once when something seems wrong, then commit if the user decides.
- Prefer plain speech over assistant-safe phrasing.
- Informal language is allowed when it fits the moment.
- Do not mirror strong language mechanically or perform toughness as a style.
- Do not retreat into sterile language to create distance.
- Stay legible: casual is fine; sloppy, evasive, or rambling is not.
Codebase (Optional)
| Path | Contents |
|---|---|
Docs Index
| Doc | Path |
|---|---|
| PRD | docs/core/PRD.md |
| Architecture | docs/core/ARCHITECTURE.md |
| Story Spec | docs/reference/STORY.md |
1. Purpose
enso is a seam-oriented harness protocol for agentic work.
A seam is a boundary where two behaviors meet and where one side can be changed without rewriting the other. Every seam has an interface (the contract that both sides agree to) and an enabling point (where you plug in a different driver). enso places intentional seams at every boundary where behavior changes hands in an agent system.
The major seams in enso:
| Seam | Interface | Enabling Point |
|---|---|---|
| Planning -> Execution | enso.story/v1 story contract (frontmatter state + required sections) |
The live story doc itself -- planned and reviewed before implementation |
| Role -> Role | Story state, role-owned sections, transition log, human gates | Planner, Generator, Evaluator, Human, and Orchestrator transitions over the same story instance |
| Ephemeral -> Persistent | Six operations (Write, Select, Probe, Compress, Isolate, Assign) | The agent's explicit choice to invoke Write instead of silently mutating context |
| Agent -> Codebase | Context Scope (Write / Read / Exclude) | The scoped file list loaded at runtime |
| Agent -> Capability | SKILL.md frontmatter (name, description, compatibility) | The scripts dropped into docs/skills/<name>/ |
| Stance -> Protocol | SOUL.md / AGENTS.md dual-document structure |
Which persona files are injected into the harness |
| Human -> Surface | Orchestration surface contract (AOS.md) | The specific tools, context, and voice configured in the harness instance |
| Self-improvement | Skill bootstrap protocol (§2.1, §7) | The agent authoring a new skill instead of rewriting its system prompt |
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 · 628 lines · 5,644 tokens per session scan A f3c8a6ca3762
enso AGENTS.md is an instructions file published in the GitHub repository usefulmove/enso (2 stars, last pushed 2mo ago), licensed MIT. It adds 5,644 tokens to every session, about $0.0282 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
learn-harness-engineering CLAUDE.md
Instructions for walkinglabs/learn-harness-engineering, covering claude.md, project overview, commands, documentation site and run lecture code examples.
trueforge AGENTS.md
AGENTS.md instructions for truefoundry/trueforge, a project described as: The open-source agent harness - the runtime layer that turns an LLM into a working agent.
aif-handoff CLAUDE.md
Instructions for lee-to/aif-handoff, covering claude.md, project overview, tech stack, project structure and key entry points.
har AGENTS.md
Instructions for os-factory/har, covering har — agent development guide, har / agent environment, harnesses in this repo, harness workflow (dogfooding) and run history.
sprawling AGENTS.md
Instructions for 2youg1/sprawling, covering agents.md — how work is done in this repository, the loop, read before you write, one change, five steps and the rules a machine holds.
opencode-plusplus AGENTS.md
Instructions for whut09/opencode-plusplus, covering agent guide, generated agent guide, must-read rules, default workflow and project entrypoints.