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 bzantium/meta-harness --skill update-harnessgit clone --depth 1 https://github.com/bzantium/meta-harnessWrote 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/bzantium/meta-harness/update-harness)<a href="https://agentmods.dev/skills/bzantium/meta-harness/update-harness"><img src="https://agentmods.dev/badge/skills/bzantium/meta-harness/update-harness/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/bzantium/meta-harness/update-harness"><img src="https://agentmods.dev/badge/skills/bzantium/meta-harness/update-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00112 | $0.02121 |
| Opus 5 | $0.00056 | $0.01060 |
| Sonnet 5 | $0.00022 | $0.00424 |
| Haiku 4.5 | $0.00011 | $0.00212 |
Grade A, and why
update-harness 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 9d 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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
update-harness — Harness Health Check & Updater
Analyzes an existing Claude Code harness against current official features, project evolution, and the pattern library. Produces actionable recommendations: what to remove, update, and add.
Workflow
Phase 1: Inventory
Scan the existing harness and catalog every component:
- Agents — Read all
.claude/agents/*.md: name, model, role, constraints - Skills — Read all
.claude/skills/*/SKILL.md: name, triggers, workflow - Hooks — Read
.claude/settings.jsonhooks section: events, matchers, scripts - Rules — Read
CLAUDE.md: delegation rules, conventions, agent catalog - MCP — Read
.mcp.jsonif present - Plugins — Read
.claude/settings.jsonenabled plugins
Produce an inventory list with component count. If no harness is found, inform the user and suggest using create-harness instead.
Phase 2: Official Features Delta
Check what the harness duplicates or could leverage from official features:
- Read
../create-harness/references/official-builtins.md - WebSearch "Claude Code changelog latest features" for anything newer than the reference
- For each harness component, evaluate:
- Redundant? — Does an official feature now do the same thing? (e.g., a custom explore agent when built-in Explore exists)
- Superseded? — Has a new official feature made this component unnecessary? (e.g., custom task tracking when TaskCreate is now built-in)
- Leverageable? — Could an official feature enhance this component? (e.g., a skill could use CronCreate instead of custom scheduling)
Phase 3: Project Evolution
Spawn scout to re-analyze the current project state:
Agent(
name: "scout",
prompt: "Analyze this project. Follow your investigation protocol and produce the structured analysis report.",
model: "sonnet"
)
Compare the scout's current findings against what the harness covers:
| Change | Implication |
|---|---|
| New modules/services added | Harness may need new agents or wider scope |
| Modules removed | Harness may reference dead code |
| Tech stack changed (new framework, build tool) | Hooks/skills may target wrong tools |
| Test framework changed | Verification hooks may be broken |
| Team size changed (solo to team or vice versa) | Harness scale may be wrong |
| New external integrations (APIs, databases) | May need safety hooks or specialized agents |
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.
- 9d ago First seen · 194 lines · 112 tokens per session scan A 08d2142011d8
update-harness is a skill published in the GitHub repository bzantium/meta-harness (2 stars, last pushed 5mo ago), licensed MIT. It adds 112 tokens to every session and 2,121 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
goal-loop
Loop engineering as gradient descent — a guided goal loop where an agent factory generates goal-specialized agents (implementer, verifier, diagnoser, judge) and iterates forward → loss → backward → update until the goal is provably met. Evidence is split into visible validation (the implementer's loss) and held-out…
metaprompt
Generate a complete, ready-to-use prompt for a target model and harness. Triggers: metaprompt, generate a prompt for, write me a prompt, create a system prompt, prompt engineer this, optimize this prompt.
file-todos
File-based TODO lifecycle for tracking review findings, managing technical debt, and organizing work items. Each TODO is a standalone markdown file with YAML frontmatter, named by convention: {id}-{status}-{priority}-{description}.md. Status transitions drive the lifecycle: pending (created from review) -> ready…
codex-judge
Cross-provider LLM judge using Codex (GPT-5) to evaluate implementations written by Claude. Internal skill invoked by the insistir lead during the review loop. Not user-facing.
agent-sdk-wizard
Build a working Claude Agent SDK agent by asking the user 9 illustrated questions, one at a time, then generating and running the code. Trigger on "crea un agente", "quiero hacer un agente con el SDK", "ayúdame a armar mi agente", "nuevo agente paso a paso", "agent sdk wizard", "build an agent with the Claude Agent…
image-to-frontend
This skill should be used when the user asks to "image to frontend", "build this UI", "make me a landing page", "design and ship this site", "generate a mockup", "turn this screenshot into code", or hands over a product brief or reference image and wants a working React or HTML page. Also covers single-shot app…