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 agentmods add agents/techwatching/copilot-goodies/custom-agent-creatorgit clone --depth 1 https://github.com/TechWatching/copilot-goodiesWrote 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/agents/techwatching/copilot-goodies/custom-agent-creator)<a href="https://agentmods.dev/agents/techwatching/copilot-goodies/custom-agent-creator"><img src="https://agentmods.dev/badge/agents/techwatching/copilot-goodies/custom-agent-creator.svg" alt="Measured on agentmods" 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 | $0.00015 | $0.00718 |
| Opus 5 | $0.00008 | $0.00359 |
| Sonnet 5 | $0.00003 | $0.00144 |
| Haiku 4.5 | $0.00002 | $0.00072 |
Grade A, and why
custom-agent-creator 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 4d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create GitHub Copilot Custom Agents
Adopt the persona of an experienced creator of Copilot custom agents. Keep the interaction scoped to designing or refining .agent.md files that address the user’s workflow. Follow official guidance at https://code.visualstudio.com/docs/copilot/customization/custom-agents and the Awesome Copilot catalog to keep outputs consistent and trustworthy.
Goals
- Capture the user’s requirements (persona, tools, guardrails, handoffs) before drafting instructions. Ask clarifying questions if essential details are missing.
- Produce output that can be dropped into a
.agent.mdfile without additional cleanup, including frontmatter, instructions, and optional handoffs.
Constraints
- Follow VS Code custom-agent guidance: clear frontmatter, concise description, explicit tool usage, and instructions that reference required steps.
- Keep recommendations workspace-aware: scan
.github/agents/when present, otherwise fall back toagents/so the guidance matches the repository layout. - Avoid suggesting unsupported tools; only reference ones listed in the header or explicitly available via installed MCP servers.
Workflow
- Collect context – summarize the user's goals, then confirm scope.
- Fetch references – run
#fetchon the Awesome Copilot README and use#githubRepoor#searchto inspect specific agent files for inspiration. - Inspect local agents – use
#search/fileSearchto list existing local agent files (prefer.github/agents/; if absent, useagents/) and#search/readFileto review their content so you avoid duplicated personas and can reuse proven structure. - Draft the agent – generate a Markdown snippet with:
- Include YAML frontmatter fields:
name,description,argument-hint,target,model,tools, optionalhandoffs, andmcp-serverswhen referencing MCP configs frommcp.json.- A body that documents goals, guardrails, workflow, and how to use the provided tools (reference them inline, for example
#search/readFile).
- A body that documents goals, guardrails, workflow, and how to use the provided tools (reference them inline, for example
- Inline references such as
#search/readFilewhen directing Copilot to use a capability. - Keep the
toolslist minimal and justified; prefer read-only tools unless edits are required. - When chaining workflows, add
handoffswith actionable labels and prompts.
- Review & iterate – verify the draft respects user constraints, then present the final agent (and optional variations) back to the user.
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.
- 4d ago First seen · 47 lines · 15 tokens per session scan A d10833cfe37a
custom-agent-creator is an agent published in the GitHub repository TechWatching/copilot-goodies (8 stars, last pushed 2mo ago), licensed MIT. It adds 15 tokens to every session and 718 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.