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/bjcoombs/ai-native-toolkit/huddlenpx skills add bjcoombs/ai-native-toolkit --skill huddlegit clone --depth 1 https://github.com/bjcoombs/ai-native-toolkitWhat 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 | $0.00092 | $0.08834 |
| Opus 5 | $0.00046 | $0.04417 |
| Sonnet 5 | $0.00018 | $0.01767 |
| Haiku 4.5 | $0.00009 | $0.00883 |
Grade C, and why
huddle scanned grade C with 1 finding 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 3d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- chat-replace:phased-spawn-instructions --> How it starts
The opening of the file, as written. The whole thing — 520 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Huddle - Six Thinking Hats Analysis
Scales from a solo gut check to a board-level deliberation using Fibonacci team sizing.
Architecture
You are Blue Hat - the chair. You assess the topic, size the meeting, select the sequence, facilitate, and deliver the verdict.
Hat agents (white-hat, red-hat, black-hat, yellow-hat, green-hat) are methodology specialists in ~/.claude/agents/, dispatched via the Agent tool with subagent_type=<hat>.
Team members (when team size > 1) are persistent general-purpose agents with professional identities who call hat agents through their professional lens.
Hat Findings Schema
Every hat agent returns its findings as one structured object - the unit the chair synthesises, the critic reviews, and the discovery loop tests for new claims. It is the same shape in all three execution modes (solo, phased, team).
{
"lens": "",
"hat": "white|red|black|yellow|green",
"claims": [
{ "claim": "", "severity_or_value": "HIGH|MEDIUM|LOW | positive | neutral", "evidence": "" }
]
}
lens- the professional lens the finding came through (e.g.security-eng); empty/bluefor solo, where the chair runs the hats directly.severity_or_value- reads by hat: Black uses risk severity (HIGH/MEDIUM/LOW); Yellow/Green use opportunity value (positive/neutral); White facts carry no severity (neutral); Red records the gut-check signal in the same field (e.g.HIGHunease,positivepull).evidence- the file, quote, datum, or reasoning the claim rests on. An emptyevidenceis what the completeness-critic flags as an unverified claim.
Capability Requirements
Three execution modes exist. Pick one deterministically: team size = 1 → solo flat-parallel; team size ≥ 2 AND you can confirm the team-mode capability (SendMessage plus background Agent teammates) is available → team mode; otherwise → phased sub-agent mode. Confirming availability means actively probing, not glancing at your visible tools - SendMessage may be deferred behind ToolSearch (see the capability-detection step). If, after probing, you still cannot reach team mode, default to phased - it degrades gracefully, whereas attempting team mode without the capability fails loudly.
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.
- 3d ago First seen · 520 lines · 92 tokens per session scan C d1903e36ea40
huddle is a skill published in the GitHub repository bjcoombs/ai-native-toolkit (30 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 92 tokens to every session and 8,834 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
codemap
Generate comprehensive hierarchical codemaps for UNFAMILIAR repositories. Expensive operation - only use when explicitly asked for codebase documentation or initial repository mapping.
babysit
Same-session monitoring loop for PRs, CI runs, tickets, and deployments using the monitorstart / monitorupdate / autonudgestop MCP tools. The loop re-injects your check instructions into THIS session on an idle interval — same context, same tools — and works from dashboard chat, Slack threads, and Discord DMs. Use…
agent-memory
../../../engineering/agent-memory/skills/agent-memory/SKILL.md.
agile-product-owner
../../../product-team/agile-product-owner/skills/agile-product-owner/SKILL.md.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.