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 Frontal-so/outbound-skills --skill claygentgit clone --depth 1 https://github.com/Frontal-so/outbound-skillsWrote 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/frontal-so/outbound-skills/claygent)<a href="https://agentmods.dev/skills/frontal-so/outbound-skills/claygent"><img src="https://agentmods.dev/badge/skills/frontal-so/outbound-skills/claygent/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/frontal-so/outbound-skills/claygent"><img src="https://agentmods.dev/badge/skills/frontal-so/outbound-skills/claygent.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 | $0.00131 | $0.01040 |
| Opus 5.5 | $0.00052 | $0.00416 |
| Sonnet 5.5 | $0.00026 | $0.00208 |
| Haiku 4.5 | $0.00013 | $0.00104 |
Grade A, and why
claygent 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 2mo 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claygent -- AI Research Agent
You help users leverage Claygent for research tasks that require web browsing and custom data extraction beyond standard enrichment providers.
Reference
Read {SKILL_BASE}/resources/prompts/claygent-guide.md for model options, credit costs, output formatting, and best practices.
When to Use Claygent vs Standard Enrichment
| Use Claygent | Use Standard Enrichment |
|---|---|
| Data not in any database | Email, phone, firmographics |
| Real-time web research needed | Static company/people data |
| Custom/nuanced questions | Industry, revenue, headcount |
| Competitor mentions, case studies | Tech stack (use HG Insights) |
| Funding news, press mentions | LinkedIn profile data |
Model Selection
- Claygent Neon (1-2 credits) -- Clay's flagship, optimized for extraction into columns. Use this 90% of the time.
- GPT-4 (2-3 credits) -- only for complex reasoning tasks
- Claude Opus (2-3 credits) -- only for complex reasoning tasks
Rule: GPT-4 Mini / Claygent Neon handles 90% of tasks. Do not overspend on advanced models.
Prompting Rules (from Eric Noski)
- One task per Claygent column -- never combine "check site + classify + extract" in one prompt
- Use "purple" as null keyword -- "If not found, output 'purple'" (easy to filter downstream)
- Include 3-4 examples in every prompt -- even mediocre prompts become excellent with examples
- Ask for reasoning before answer on complex tasks -- improves accuracy significantly
- 10-minute manual research rule -- research 3-4 records manually first, then automate that exact process
Prompt Template
Visit /company_domain and answer: [SPECIFIC QUESTION]
Rules:
- Only use information found on the website
- If not found, output "purple"
- Output format: [text/number/true-false/URL]
Example outputs:
- "Series B, $25M from Sequoia"
- "purple"
Credit Optimization
- Simple yes/no questions: 1 credit
- Deeper analysis: 2-3 credits
- Iterate in the builder (free) before running on full table
- Add conditional run: only run Claygent on rows where standard enrichment left gaps
- Never use Claygent for data available via standard providers (email, phone, revenue)
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.
- 2mo ago First seen · 87 lines · 131 tokens per session scan A d9b4a377f300
claygent is a skill published in the GitHub repository Frontal-so/outbound-skills (9 stars, last pushed 1mo ago), licensed MIT. It adds 131 tokens to every session and 1,040 once invoked, about $0.0005 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 2026-08-31.
claygent compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| claygentFrontal-so/outbound-skills | Skill | 9 | 1mo ago | MIT | A | 131 | Claude Code |
| audience-icp-filterLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 228 | Claude Code, Codex |
| team-performance-dashboardLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 235 | Claude Code, Codex |
| weekly-performance-advisorLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 215 | Claude Code, Codex |
| outreach-icp-finderLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 234 | Claude Code, Codex |
| objection-analyzerLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 215 | Claude Code, Codex |
| sales-nav-search-builderLaGrowthMachine/gtm-system | Skill | 46 | 1mo ago | MIT | A | 221 | Claude Code, Codex |