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 scolladon/craft --skill craft-rungit clone --depth 1 https://github.com/scolladon/craftWrote 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/scolladon/craft/craft-run)<a href="https://agentmods.dev/skills/scolladon/craft/craft-run"><img src="https://agentmods.dev/badge/skills/scolladon/craft/craft-run.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 5.1 | $0.00033 | $0.00350 |
| Opus 5 | $0.00016 | $0.00175 |
| Sonnet 5 | $0.00007 | $0.00070 |
| Haiku 4.5 | $0.00003 | $0.00035 |
Grade A, and why
craft-run 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 7d 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.
What it actually says
craft — Antigravity entrypoint
The full orchestrator procedure lives in ${CRAFT_ROOT}/skills/run/SKILL.md. Read it
and follow it as the orchestrator — this file only records the Antigravity-specific
binding notes; it does not restate the procedure.
Binding notes (measured, not assumed — see docs/contributing/specs/antigravity-poc-record.md):
- Single-agent topology. The Antigravity (Gemini) agent IS both orchestrator and
worker. The role agents in
agents/craft-*.mdare followed as self-directed guidance per phase, NOT spawned as parallel subagents — Antigravity exposes no pinned headless spawn verb, so the multi-agent fan-out other bindings use is degraded to one agent here. - Run the engine via
run_command. Each engine step isnode ${CRAFT_ROOT}/engine/bin/<bin>.js …executed through the agent'srun_commandtool; keepCRAFT_ROOTset to the craft checkout. - The guard is on.
plugins/craft/hooks.jsonregisters a PreToolUse hook that deniesgit diff/git showwithout--no-ext-diff; keep such commands compliant. - No PR automation is bound. Push/PR steps run as ordinary
run_commandgit/gh calls under your own review, not through a bound delivery port.
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.
- 7d ago First seen · 25 lines · 33 tokens per session scan A 616f0330bf79
craft-run is a skill published in the GitHub repository scolladon/craft (2 stars, last pushed 19d ago), licensed MIT. It adds 33 tokens to every session and 350 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
terry
Invoke when orchestrating a FLEET of parallel long-running agent tasks via a priority-queue manager over git worktrees, OR when you want a producer-consumer task queue for agents that caps concurrency to avoid OOM and pace token spend on a budget — agent queue + task list + dispatched agents + a self-pacing heartbeat…
sdlc-accelerate
End-to-end SDLC ramp-up from idea to construction-ready with automated phase transitions.
project-awareness
Detect project type, AIWG framework state, team configuration, and active work to summarize status and recommend next actions.
workspace-prune-working
Clean up .aiwg/working/ by promoting, archiving, or deleting temporary files.
artifact-metadata
Manage artifact metadata, versioning, ownership, and review history across the SDLC lifecycle.
migrate-hook
Migrate existing CLAUDE.md full-injection to the AIWG.md hook file architecture.