Deep Agents Core
langchain-ai/langchain-skills
Explains how to build agents with the Deep Agents framework: create_deep_agent, the built-in middleware, the harness, SKILL.md format and configuration options.
Designs, evaluates and improves the harness around an AI agent: tool permissions, approval gates, state, memory, evals and observability, with phased plans.
Install Agentic Harness Design and Review
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnesses --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/n-agentic-harnesses .claude/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .claude/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnessesType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnesses --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/n-agentic-harnesses .agents/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .agents/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnesses --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/n-agentic-harnesses .cursor/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .cursor/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/NateBJones-Projects/OB1.git --path skills/n-agentic-harnesses--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnesses --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/n-agentic-harnesses .gemini/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .gemini/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnessesInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/n-agentic-harnesses .github/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .github/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install NateBJones-Projects/OB1 n-agentic-harnesses --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NateBJones-Projects/OB1.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/n-agentic-harnesses .opencode/skills/n-agentic-harnesses && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "n-agentic-harnesses" agent skill from https://github.com/NateBJones-Projects/OB1/tree/main/skills/n-agentic-harnesses into .opencode/skills/n-agentic-harnesses/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "n-agentic-harnesses", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
n-agentic-harnesses- GitHub stars
- 4.7k
- Token cost
- ~1.8k tokens
- SKILL.md length
- 790 words
- Files
- 16 (incl. references)
- Skills in repo
- 24
- Repo updated
- First seen
- Licence
- Custom licence
At a glance
Designs, evaluates and improves the harness around an AI agent: tool permissions, approval gates, state, memory, evals and observability, with phased plans.
- Works in 4 steps: Gather Context → Classify The Request → Classify The Product Shape → …
- Designing the tool-use and permission layer for a new agent
- SKILL.md covers Problem, Trigger Conditions, Default Posture and Step 0: Gather Context, plus 6 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
The skill treats most AI product failures as harness problems rather than model weakness: unclear tool boundaries, missing approval policy, brittle state, sloppy context assembly, no evaluation loop and little operator visibility. It triggers on design or rebuild requests and on symptoms such as tools firing without permission, sessions dying on a crash, stale context and unexplained costs.
The agent first gathers context, or reads the real codebase, configs and hooks when evaluating, then picks a design or evaluation mode and loads matching files from a numbered reference set covering principles, architectures, tool permissions, state and durability, memory and evaluation, extensibility, UX and observability, plus build and improvement playbooks. Defaults favor lean, solo-maintainable, single-agent designs, an evaluation plan even for new builds, and advice turned into phases, success criteria and failure tests.
When your agent uses it
- Designing the tool-use and permission layer for a new agent
- Reviewing an existing agent for missing approval gates or durability gaps
- Planning memory, state and resumability for long-running sessions
- Building an evaluation and observability plan for an AI product
Example prompts
- “Design the harness for a support copilot that can issue refunds, with approval gates and an eval plan.”
- “Evaluate our agent's codebase for session persistence and retry problems.”
- “Our agent costs keep climbing and nobody can see why, so find the harness gaps.”
Workflow steps
4 steps, taken from the step headings in SKILL.md.
- Gather Context
- Classify The Request
- Classify The Product Shape
- Read The Smallest Useful Reference Set
What it can do on your machine
Read from SKILL.md and the folder at commit 238df6c. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Agentic Harness Design and Review loads about 1.8k tokens when it runs, and up to ~11k if it reads all its reference files. Until then it costs about 144 tokens; SKILL.md has 790 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 790 words (~1,833 tokens).
“Most AI products do not break because the model is too weak. They break at the harness layer: unclear tool boundaries, missing approval policy, brittle state, sloppy context assembly, no evaluation loop, and weak operator visibility. This skill turns those…”
- name
- n-agentic-harnesses
- author
- Jonathan Edwards
- version
- 1.0.0
Files
SKILL.md and 15 other files (references) in skills/n-agentic-harnesses of NateBJones-Projects/OB1.
- SKILL.md
- README.md
- agents/openai.yaml
- metadata.json
- references/01-principles-and-solo-dev-defaults.md
- references/02-harness-shapes-and-architecture.md
- references/03-tools-execution-and-permissions.md
- references/04-state-sessions-and-durability.md
- references/05-context-memory-and-evaluation.md
- references/06-agents-and-extensibility.md
- references/07-ux-observability-and-operations.md
- references/08-design-and-build-playbook.md
- references/09-evaluation-and-improvement-playbook.md
- references/10-example-requests-and-output-patterns.md
- references/11-codex-translation-notes.md
- variants
Open the folder on GitHubat commit 238df6c
Compare with similar skills
Agentic Harness Design and Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Agentic Harness Design and Review this skillNateBJones-Projects/OB1 | 4.7k | — | ~1.8k | Automated safety check: Pass | Custom licence | |
| Deep Agents Corelangchain-ai/langchain-skills | 1.3k | — | ~3.1k | Automated safety check: Pass | MIT | |
| Dive Into LangGraphluochang212/dive-into-langgraph | 457 | — | ~837 | Automated safety check: Notes | Custom licence | |
| Create Agentvectorize-io/hindsight | 48k | — | ~1.1k | Automated safety check: Pass | MIT | |
| Neo4j Agent Memory Skillneo4j-contrib/neo4j-skills | 114 | — | ~5.8k | Automated safety check: Pass | MIT | |
| Agent Developmentsundial-org/awesome-openclaw-skills | 663 | — | ~2.4k | Automated safety check: Pass | MIT |
Similar skills
- Official
Dive Into LangGraph
luochang212/dive-into-langgraph
A Chinese-language guide and reference for building agents with LangGraph 1.0, from a first ReAct agent through middleware, memory, MCP, RAG and web search.
Create Agent
vectorize-io/hindsight
Create a new Hindsight-powered subagent with long-term memory.
Neo4j Agent Memory Skill
neo4j-contrib/neo4j-skills
Authoritative reference for the neo4j-agent-memory Python package — a graph-native memory system for AI agents built on Neo4j — and for the hosted service (NAMS) at memory.neo4jlabs.com.
Agent Development
sundial-org/awesome-openclaw-skills
Design and build custom Claude Code agents with effective descriptions, tool access patterns, and self-documenting prompts.
Mem0 Platform SDK
mem0ai/mem0
Adds persistent memory to AI apps with the Mem0 Python and TypeScript SDKs: store, search, update and delete user memories, with framework integrations.
More from NateBJones-Projects/OB1
All 24 skills in this repoHeavy File Ingestion
NateBJones-Projects/OB1
Converts large PDF, DOCX, PPTX, XLSX and CSV files into markdown or CSV plus an index before the agent reads them, so tokens go to the compressed copy.
Aiception Skill Extraction
NateBJones-Projects/OB1
Pulls reusable knowledge out of work sessions and turns it into new skills, checking existing notes and skills first to avoid duplicates.
Claude Code Auto-Capture Hook
NateBJones-Projects/OB1
Fires a Stop hook that captures a Claude Code session transcript automatically when the session ends without a verbal wrap-up.
Open Brain Local HTTP
NateBJones-Projects/OB1
Captures and searches personal notes in a self-hosted Open Brain using plain curl calls over HTTP, for setups where MCP is disabled or blocked.
Panning for Gold
NateBJones-Projects/OB1
Processes voice transcripts and brain dumps in three phases, extracting every idea thread, evaluating the strongest and saving a permanent inventory and synthesis.
Weekly Signal Diff
NateBJones-Projects/OB1
Turns a noisy week of AI or software market news into a short list of structural changes, weighted by what the user already tracks in Open Brain memory.
Related
Categories
Questions about Agentic Harness Design and Review
What does Agentic Harness Design and Review do?
Designs, evaluates and improves the harness around an AI agent: tool permissions, approval gates, state, memory, evals and observability, with phased plans. The skill treats most AI product failures as harness problems rather than model weakness: unclear tool boundaries, missing approval policy, brittle state, sloppy context assembly, no evaluation loop and little operator visibility. It triggers on design or rebuild requests and on symptoms such as tools firing without permission, sessions dying on a crash, stale context and unexplained costs.
When should I use Agentic Harness Design and Review?
Agentic Harness Design and Review fits situations like: designing the tool-use and permission layer for a new agent; reviewing an existing agent for missing approval gates or durability gaps; planning memory, state and resumability for long-running sessions; building an evaluation and observability plan for an AI product.
How do I install Agentic Harness Design and Review in Claude Code?
Run `npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a claude-code`. Or copy the skill folder (skills/n-agentic-harnesses in NateBJones-Projects/OB1) into .claude/skills/n-agentic-harnesses in your project. Claude Code loads it when a task matches its description.
How do I install Agentic Harness Design and Review in Codex?
Run `npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a codex`. Or copy the skill folder (skills/n-agentic-harnesses in NateBJones-Projects/OB1) into .agents/skills/n-agentic-harnesses in your project. Codex loads it when a task matches its description.
Can I use Agentic Harness Design and Review in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add NateBJones-Projects/OB1 --skill n-agentic-harnesses -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/n-agentic-harnesses, .gemini/skills/n-agentic-harnesses, .github/skills/n-agentic-harnesses and .opencode/skills/n-agentic-harnesses in your project.
What does Agentic Harness Design and Review need to run?
SKILL.md names no scripts, command-line tools or credentials: Agentic Harness Design and Review is instructions for the agent only.
Does Agentic Harness Design and Review access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Agentic Harness Design and Review safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Agentic Harness Design and Review use?
Agentic Harness Design and Review has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
How many tokens does Agentic Harness Design and Review use?
About 1.8k tokens (SKILL.md is roughly 7.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 9k tokens, read only when the agent opens those files.
What are the alternatives to Agentic Harness Design and Review?
Skills that share tags, products or a category with Agentic Harness Design and Review: Deep Agents Core (langchain-ai/langchain-skills, 1.3k stars), Dive Into LangGraph (luochang212/dive-into-langgraph, 457 stars), Create Agent (vectorize-io/hindsight, 48k stars) and Neo4j Agent Memory Skill (neo4j-contrib/neo4j-skills, 114 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Agentic Harness Design and Review?
NateBJones-Projects (a GitHub organization) maintains it in NateBJones-Projects/OB1, which has 4,710 GitHub stars. The repository holds 24 skills in this directory. The repository was last updated on October 9, 2026.
Source: NateBJones-Projects/OB1 on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.