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 skills add burin-labs/harn --skill harn-mcpgit clone --depth 1 https://github.com/burin-labs/harnWrote 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/skills/burin-labs/harn/harn-mcp)<a href="https://agentmods.dev/skills/burin-labs/harn/harn-mcp"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-mcp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/burin-labs/harn/harn-mcp"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 71 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 5.1 | $0.00027 | $0.01481 |
| Opus 5 | $0.00014 | $0.00740 |
| Sonnet 5 | $0.00005 | $0.00296 |
| Haiku 4.5 | $0.00003 | $0.00148 |
Grade A, and why
harn-mcp 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 10d 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Build MCP clients and servers in Harn
Use this skill when a Harn program talks to an MCP server or serves one. Pair it with [[harn-testing]] for deterministic fixtures, [[harn-agent]] for tool exposure inside an agent loop, and [[harn-providers]] when sampling routes back to a model.
Maintainers changing Harn's own MCP implementation should read
docs/src/dev/mcp-maintenance.md instead. This skill covers writing MCP
clients and servers in Harn, not editing the runtime that carries them.
Start here
docs/src/mcp-and-acp.mdis the public client and server reference.docs/src/mcp-server.mdcovers the orchestrator server,harn mcp serve.- Harn's stdio servers accept the released
initializelifecycle and the 2026-07-28 request-metadata lifecycle. Streamable HTTP uses 2026-07-28. - The official Rust SDK owns protocol mechanics; Harn owns product policy.
- Route every capability through
harness.tools.*.
Connect as a client
Connect explicitly when the server is chosen at runtime:
const client = harness.tools.mcp_connect("npx", ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"])
const info = harness.tools.mcp_server_info(client)
const tools = harness.tools.mcp_list_tools(client)
const content = harness.tools.mcp_call(client, "read_file", {path: "/tmp/data.txt"})
harness.tools.mcp_disconnect(client)
Resources and prompts use mcp_list_resources, mcp_read_resource,
mcp_list_resource_templates, mcp_list_prompts, and mcp_get_prompt.
mcp_callreturns a string for a single text result, a list of content dicts for multiple blocks, or nil when empty.mcp_callthrows when the tool reports an error. Catch it where the pipeline can recover; do not let a remote failure read as an empty result.- Disconnect what you connect. Prefer declared servers when the set is fixed.
Declare servers instead of connecting by hand
Put a fixed server set in harn.toml as repeated mcp array-of-tables
entries. Each entry carries name, plus either command and args for stdio,
or transport = "http" and url. Declared servers connect before the pipeline
runs and appear in the global mcp dict.
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.
- 10d ago First seen · 139 lines · 27 tokens per session scan A d60565d9ea21
harn-mcp is a skill published in the GitHub repository burin-labs/harn (21 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 1,481 once invoked, about $0.0001 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 skills, from other repositories
agent-communication-protocol
Open protocol for AI agent interoperability enabling standardized communication between agents, applications, and humans across different frameworks.
archestra-dev-interactions-migrations
Use BEFORE writing or running any Drizzle migration that touches the interactions table (or any other very large, write-hot table). The interactions table is the platform's biggest, append-heavy table — every LLM proxy call writes a row — so a careless migration can take a write-blocking lock and stall the proxy.…
archestra-dev-llm-providers
Use when adding an LLM provider, changing proxy adapters or provider routes, fixing streaming/tool-call translation bugs, editing model fetchers or model handling, or touching provider credentials/enums and model constants.
archestra-dev-backend
Use when adding or changing Archestra backend routes, models, services, API request/response schemas, endpoint permissions, or OpenAPI/codegen for the generated API client.
translate
A translation tool for converting user-provided text between languages, with Chinese and English as the default pair.
cqrs-implementation
Implement Command Query Responsibility Segregation for scalable architectures. Use when separating read and write models, optimizing query performance, or building event-sourced systems.