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 skills/conorbronsdon/agent-skill-builder/agent-skill-buildernpx skills add conorbronsdon/agent-skill-builder --skill agent-skill-buildergit clone --depth 1 https://github.com/conorbronsdon/agent-skill-builderWrote 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/conorbronsdon/agent-skill-builder/agent-skill-builder)<a href="https://agentmods.dev/skills/conorbronsdon/agent-skill-builder/agent-skill-builder"><img src="https://agentmods.dev/badge/skills/conorbronsdon/agent-skill-builder/agent-skill-builder.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 | $0.00037 | $0.01109 |
| Opus 5 | $0.00018 | $0.00554 |
| Sonnet 5 | $0.00007 | $0.00222 |
| Haiku 4.5 | $0.00004 | $0.00111 |
Grade A, and why
agent-skill-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 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Skill Builder
Produce a ready-to-use skill from a plain-language request. Preserve the user's runtime, scope, and requested level of implementation.
Select the mode
Parse $ARGUMENTS; default to new when it contains a description.
new <description>: create or draft a skill.review <path>: report actionable findings and run validation. Do not rewrite unless asked.migrate <path>: convert a legacy.claude/commands/*.mdfile into a skill directory while preserving behavior.
Establish the contract
Infer what is already clear. Ask only when a missing choice would materially change the result.
- Choose the runtime:
- Claude Code (default): read the Claude Code reference.
- Portable Agent Skills: read the portable reference and omit Claude-only fields.
- Read the host repository's applicable instruction files (
AGENTS.md,CLAUDE.md, or equivalents) and inspect existing skills. Match local conventions and avoid duplicate capabilities. - Identify inputs, outputs, required resources, and the destination directory.
- Classify the skill:
- Reference: standing domain rules or conventions.
- Task: a bounded workflow with a defined result and stopping condition.
- Tool-backed: a task that benefits from repeatable scripts or narrowly scoped tool grants.
Design decisions
Include only fields and instructions that change behavior.
- Name: lowercase letters, digits, and hyphens. Portable skills require the frontmatter name to match the directory.
- Description: lead with the capability and include phrases users would naturally say. Keep model-visible descriptions near 250 characters when practical.
- Invocation (Claude Code only):
- Use
disable-model-invocation: truewhen timing must remain user-controlled, especially publishing, deploying, sending, deleting, or other consequential mutations. - Use
user-invocable: falsefor background knowledge users should not run directly. - Leave both enabled when natural-language discovery is useful. Invocation control never replaces approval immediately before an external or hard-to-reverse mutation.
- Use
- Arguments (Claude Code only): pair
argument-hintwith$ARGUMENTS, positional$N, or names declared inarguments:. - Forking (Claude Code only): use
context: forkfor self-contained tasks. Addbackground: falseonly when the invoking turn needs the result immediately. - Tool grants:
allowed-toolspre-approves tools; it does not restrict the remaining pool. Grant only commands the workflow actually needs. - Dynamic injection (Claude Code only): use
!commands for bounded read-only context. Never interpolate skill arguments into shell commands, and never perform mutations during preprocessing. - Resources: keep shared instructions in
SKILL.md; put conditional details inreferences/, repeated deterministic mechanics inscripts/, and output material inassets/.
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 · 83 lines · 37 tokens per session scan A c087af1bdacf
agent-skill-builder is a skill published in the GitHub repository conorbronsdon/agent-skill-builder (4 stars, last pushed 16d ago), licensed MIT. It adds 37 tokens to every session and 1,109 once invoked, about $0.0002 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
exploratory-data-analysis
Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
golden-rss
Use when testing the rss golden build.
data-charts-tako
Search and visualize the world's data - get charts, insights, and embeddable knowledge cards for finance, economics, demographics, sports, and more.
apollo-lead-finder
Two-phase Apollo.io prospecting: free People Search to discover ICP-matching leads, then selective enrichment to reveal emails/phones (credits per contact). Creates Apollo lists. Deduplicates against existing contacts by LinkedIn URL.
google-ads-audit
Google Ads account audit and business context setup. Run this first — it gathers business information, analyzes account health, and saves context that all other ads skills reuse. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should…
webhook-management
Configure and validate CCAM webhook targets across supported chat, incident, automation, and generic providers. Use when listing provider requirements, creating or updating a target, scoping it to alert rules, sending a test notification, reviewing delivery history, or deleting a target.