agent-harness instructions files

93 tagged agent-harness, measured the same way as everything else here.

Browse within: coding-agents 14ai-coding 9agentic-workflow 6agentic-workflows 6coding-agent 6agent-framework 5agent-orchestration 5

arnesto AGENTS.md

49

saski/arnesto

Instructions file CodexOpenCode

AGENTS.md instructions for saski/arnesto, a project described as: Arnesto — the agent harness that got out of hand.

not rated 5 9d ago A 7 tokens original Unlicense

Harneloop AGENTS.md

50

Ker102/Harneloop

Instructions file CodexOpenCode

Instructions for Ker102/Harneloop, covering agent instructions for harneloop, current stage, first onboarding step, product principles and engineering defaults.

not rated 4 1mo ago A 1,014 tokens original Apache-2.0

siddhartha-yz/dsh-mcp-gateway

Instructions file CodexOpenCode

AGENTS.md instructions for siddhartha-yz/dsh-mcp-gateway, covering repository architecture contract, canonical data path, responsibility split, chatgpt and this repository.

not rated 4 3d ago A 982 tokens original MIT

vennie CLAUDE.md

52

mindtheproduct/vennie

Instructions file

Claude Code instructions for mindtheproduct/vennie, covering vennie - your product career partner, making it personal, first-time setup, user profile and active persona system.

not rated 3 4mo ago A 10,210 tokens

cc-skills CLAUDE.md

53

hssh8917/cc-skills

Instructions file

Claude Code instructions for hssh8917/cc-skills, covering claude.md, project overview, project structure, agent skills specification and frontmatter.

not rated 2 2d ago C 8,575 tokens original MIT

the-sid-dani/second-brain-os

Instructions file

Instructions for the-sid-dani/second-brain-os, covering second-brain os — project instructions for, load order at session start, configuration — source of truth for skills, workspace and templates (under / /templates/).

not rated 2 29d ago A 3,397 tokens original MIT

Oficina AGENTS.md

55

JoaoEquer/Oficina

Instructions file CodexOpenCode

AGENTS.md instructions for JoaoEquer/Oficina: Entry point for harnesses that read AGENTS.md (Codex, Cursor, OpenCode, Gemini CLI and similar).

not rated 2 changed today A 662 tokens original MIT

ouroboros AGENTS.md

56

ourostack/ouroboros

Instructions file CodexOpenCode

AGENTS.md instructions for ourostack/ouroboros, covering philosophy, constitutional commitments, delivery rules, safety rules and human override.

not rated 2 today A 4,754 tokens original MIT

mochow13/keen-agent

Instructions file CodexOpenCode

Instructions for mochow13/keen-agent, covering keen agent, important guidelines, architecture, permission system and cast to gif conversion.

not rated 2 12d ago A 317 tokens copy · 92% MIT

grimoire AGENTS.md

58

hon454/grimoire

Instructions file CodexOpenCode

Instructions for hon454/grimoire, covering grimoire agent protocol, repository intent, working rules, pull requests and maintenance policy.

not rated 2 17d ago A 298 tokens original MIT

pie AGENTS.md

59

dineshdb/pie

Instructions file CodexOpenCode

Instructions for dineshdb/pie, covering testing, architecture, i'm an idiot, simplification and scope.

not rated 2 3mo ago A 453 tokens original MIT

alexsds/ade-workflow

Instructions file

Claude Code instructions for alexsds/ade-workflow, covering ade — agent-driven engineering, what this is, architecture, development and key principles.

not rated 1 5mo ago A 666 tokens original MIT

dotoricode/tink-harness

Instructions file CodexOpenCode

Repository instructions for Tink Harness, a tool that installs and maintains commands and skills for coding assistants.

not rated 1 1mo ago A 699 tokens original MIT

elvismusli/onlyharness

Instructions file CodexOpenCode

AGENTS.md instructions for elvismusli/onlyharness, covering superskill agent guide, what this repo is, dev commands, agent use and superskill local/internal-alpha contract.

not rated 0 1mo ago A 6,091 tokens

frankxai/agentic-ops-hub

Instructions file CodexOpenCode

AGENTS.md instructions for frankxai/agentic-ops-hub, covering agents.md, unified agent configuration — single source of truth, llm behavioral guardrails (top thinkers system), 1. think before coding (karpathy rules) and 2. feynman alignment protocol (explain simply).

not rated 0 yesterday A 905 tokens