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 dryvist/claude-code-plugins --skill agent-buildergit clone --depth 1 https://github.com/dryvist/claude-code-pluginsWrote 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/dryvist/claude-code-plugins/agent-builder)<a href="https://agentmods.dev/skills/dryvist/claude-code-plugins/agent-builder"><img src="https://agentmods.dev/badge/skills/dryvist/claude-code-plugins/agent-builder/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/dryvist/claude-code-plugins/agent-builder"><img src="https://agentmods.dev/badge/skills/dryvist/claude-code-plugins/agent-builder.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 | $0.00056 | $0.01705 |
| Opus 5.5 | $0.00022 | $0.00682 |
| Sonnet 5.5 | $0.00011 | $0.00341 |
| Haiku 4.5 | $0.00006 | $0.00170 |
Grade A, and why
agent-builder 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 today.
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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent builder
Produce one canonical agent package in dryvist/ai-llm-prompts/agents/, or
the authorized private twin. Keep private content in the private repository.
This skill supplies authoring instructions, never a second agent definition.
Establish the contract
Read the target repository's instructions, current v2 schema, standards, connector roster, template inputs, evaluation runner, and import guide. Search for an existing agent or native feature that already performs the job. Extend it when appropriate; add no custom script where an existing tool works. Use [plugin-dev:agent-development][agent-development] for role design, while retaining the platform YAML format rather than a harness-specific definition.
Record the job, inputs, output schema, completion criterion, permitted tools, and authority boundary. Identify the authorized worktree and paths, network and service permissions, required checks, stop conditions, and delivery format. Proceed within those boundaries. Stop for a conflicting contract, missing schema/template/connector, missing authority, or a check failing twice after a real fix attempt. Report the specific blocker without inventing defaults.
Author the canonical inputs
Create or update agents/<name>.yaml, its agents/prompts/ prompt, and its
evaluation inputs. Take field types and allowed values from the current schema;
retain other required fields, including tool allowlists and routing roles.
| v2 field | Required decision |
|---|---|
tier |
0 interactive, 1 event-driven, or 2 batch. |
triggers |
Calls, supported events, or schedules with filters and input contracts. |
privacy_class |
Schema-approved class and permitted data destinations; private work stays local. |
connectors |
Named connector agents from the roster, with only the operations needed. |
code_work |
Declare whether code work is required; all edits, git, and tests run in agent-sandbox jobs. |
mode |
Start in shadow; live requires the promotion evidence below. |
eval_set |
Existing runner's path to this agent's versioned regression suite. |
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.
- today First seen · 147 lines · 56 tokens per session scan A 9de8f25df985
agent-builder is a skill published in the GitHub repository dryvist/claude-code-plugins (3 stars, last pushed today), licensed Apache-2.0. It adds 56 tokens to every session and 1,705 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-06.
agent-builder compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agent-builderdryvist/claude-code-plugins | Skill | 3 | today | Apache-2.0 | A | 56 | Claude Code, Codex |
| ai-ml-developmenttravisjneuman/.claude | Skill | 107 | 2d ago | MIT | A | 43 | Claude Code, Codex |
| llm-app-developmenttravisjneuman/.claude | Skill | 107 | 2d ago | MIT | A | 22 | Claude Code, Codex |
| ai-policy-generatortravisjneuman/.claude | Skill | 107 | 2d ago | MIT | A | 42 | Claude Code, Codex |
| data-sciencetravisjneuman/.claude | Skill | 107 | 2d ago | MIT | A | 47 | Claude Code, Codex |
| Vizra ADK Memory Systemvizra-ai/vizra-adk | Skill | 296 | 28d ago | MIT | A | 24 | Claude Code, Codex |
| data-engineeringtravisjneuman/.claude | Skill | 107 | 2d ago | MIT | A | 70 | Claude Code, Codex |