AI & LLM Skill
Agent Eval Suite Langsmith
Production agent eval suite LangSmith dataset curation + Promptfoo assertion framework +…
$15
Claude Agent Template Library
A categorized canon of 100+ Claude Code subagent templates with a strict frontmatter standard, enforcing Mission Brief, Agent Chain and bypass-permissions discipline.
A categorized canon of 100+ Claude Code subagent templates with a strict frontmatter standard (name, description, tool whitelist, model alias, system prompt). It enforces the three disciplines that make agent dispatch reliable: the Mission Brief pattern, Agent Chain handoffs, and the never-forgotten bypass-permissions mode. Build new agents from proven templates instead of from scratch, and retroactively audit your existing fleet for missing guardrails.
$15 one-time
Add to a kit → Prices include 20% VAT. · Forged on real agency work · one-time, no lock-in
- Type Skill
- Category AI & LLM
- Delivery Email · instant
- License One-time
Inside the run · no black box
See the actual work before you buy it.
New agents start from a 100+ template taxonomy, not a blank file. Cloning, strict frontmatter, a Does and Does-not table and a mandatory downstream chain keep every agent single-purpose and auditable.
- Classifies the requested agent against a 100+ template taxonomy (research, planning, coding, writing, multi-agent, RAG, browser automation) and clones the closest canonical template instead of writing from scratch
- Fills the fixed frontmatter skeleton: name, trigger-keyword description for intent matching, model alias instead of a hardcoded version, a pattern-based tool whitelist (never open-ended shell access), and chain links to predecessor and successor agents
- Writes the Does / Does-not table so the agent carries a single responsibility, plus an anti-pattern section of at least five forbidden behaviors
- Embeds a Mission Brief parser into the system prompt: the agent must reject any dispatch missing purpose, in-scope, out-of-scope, done criteria or verification, which kills scope creep at the door
- Runs the audit script across all existing agents, checking frontmatter completeness, chain discipline and tool whitelist strictness, and outputs a PASS / NEEDS-PATCH table per agent
- Registers the mandatory next-hop chain (code work hands off to QA, content hands off to SEO review) so no agent output ships without its downstream check
One power source. 6 lines out.
claude-agent-template-library · core
core active · 6 lines
- ✓ authoring a new subagent
Authoring a new subagent with correct frontmatter and tool whitelist
- ✓ adding a new capability
Adding a new capability to an existing agent without breaking its scope
- ✓ designing a multi-agent
Designing a multi-agent pipeline with mandatory chain handoffs
- ✓ creating a platform-spec…
Creating a platform-specific specialist agent for a new customer type
- ✓ auditing your current ag…
Auditing your current agent fleet for missing guardrails
- ✓ debugging dispatch failu…
Debugging dispatch failures from a forgotten permissions mode
Yours to keep.
Drag time forward. Watch what stays.
Forever
That's what owning means.
The rented stack
ai writing tool: subscription
expired · access lostanalytics suite: subscription
expired · access lostdesign platform: subscription
expired · access lost(nothing left)
Your forge
-
Save 80%+ effort by cloning canon templates instead of writing agents from zero
license: perpetual -
Eliminate dispatch interruptions by enforcing bypass-permissions mode every time
license: perpetual -
Keep each agent focused with single-responsibility 'does / doesn't' tables
license: perpetual -
Prevent scope creep with the 5-field Mission Brief on every dispatch
license: perpetual
subscriptions expire · deeds don't
Everything in the box.
Pick a piece up. Watch it work.
A complete agent frontmatter skeleton with tool pattern-whitelisting and model alias rules
6 parts · one working system · ships instantly by email
This wasn't forged for everyone.
- Not for you if you'd rather rent a tool than own one.
- Not for you if you want someone else to run your stack.
- Not for you if you're happy guessing.
Teams building and maintaining a fleet of Claude Code subagents who want reliable, standardized dispatch instead of one-off agents that need constant supervision.
then this was forged for you.Works with
Universal by design: these run in any AI. Delivered in the open Agent Skills + MCP format (native in Claude); ChatGPT, Gemini, Cursor and Copilot adapt the same files their own way.
- Claude Native format
- ChatGPT Adapts via open standards
- Gemini Adapts via open standards
- Cursor Adapts via open standards
- Copilot Adapts via open standards
Catch what's on your mind.
the air is clear. nothing between you and the forge.
catch a spark: the forge will answer
-
Are these tied to Claude Code, or generic agent prompts?
They are built around the Claude Code subagent format, with the frontmatter fields dispatch actually reads: name, description, tool whitelist, model alias and system prompt. That is what lets them drop in rather than need translation.
-
Is a library of a hundred-plus templates just a pile to dig through?
They are categorized as a canon, and the value is the shared standard underneath: the Mission Brief pattern, chain handoffs and the dispatch settings that are easy to forget. You copy a known-good shape instead of re-deriving it each time.
-
Will a template give my agent its actual domain expertise?
No, it gives you a correct, scoped skeleton with the disciplines wired in. The specific instructions and knowledge for your task are still yours to write into the system prompt.
-
How is it delivered?
By email right after purchase: ready to run, downloaded instantly, no setup wait.
-
One-time or subscription?
A one-time purchase; no subscription or hidden fees. VAT (20%) is included.
-
Can I get a refund?
As a digital product, it can’t be refunded once downloaded. That’s why we show exactly what’s inside and who it’s for, right here.