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 ashfulcra/fulcra-tools --skill fulcra-agent-atcgit clone --depth 1 https://github.com/ashfulcra/fulcra-toolsWrote 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/ashfulcra/fulcra-tools/fulcra-agent-atc)<a href="https://agentmods.dev/skills/ashfulcra/fulcra-tools/fulcra-agent-atc"><img src="https://agentmods.dev/badge/skills/ashfulcra/fulcra-tools/fulcra-agent-atc/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ashfulcra/fulcra-tools/fulcra-agent-atc"><img src="https://agentmods.dev/badge/skills/ashfulcra/fulcra-tools/fulcra-agent-atc.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00091 | $0.04767 |
| Opus 5 | $0.00046 | $0.02384 |
| Sonnet 5 | $0.00018 | $0.00953 |
| Haiku 4.5 | $0.00009 | $0.00477 |
Grade A, and why
fulcra-agent-atc 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 2d 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 — 324 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fulcra Agent ATC
A subscription fleet — Claude Max, OpenAI/Codex plans, maybe a local ollama and a hosted overflow lane, across one or several harnesses — has two failure modes that waste money and stall work: it burns frontier budget on mechanical work a cheap model would finish, and it stalls on one capped account — hitting a window cap on one subscription while another sits idle. ATC (air-traffic control) is the routing layer that fixes both, and it runs on one account the same way it runs on a fleet.
Three parts:
- The map — a packaged default model map (
map_version 2026-07-08) tagging every model by capability (code,architecture,writing,long-context,vision,fast,tool-use) and a cost rank (1 = scarcest cap-weight, 9 = locally free). You declare accounts; the map says which models each can run and what they cost. - The router —
coord-engine route <team> --needs …: ranks the models that cover your declared needs cheapest-capable-first, filtered to the accounts that have headroom right now.coord-engine headroomis the shared cap ledger it reads. - The feedback loop —
coord-engine usage log … --model --task-class --outcome: every dispatch logs what it cost and how it turned out. Repeated bad outcomes on a (model, task_class) pair demote it — the router marks it and ranks it below everything that hasn't been failing.
Dispatch itself stays native to your harness — ATC decides which model against which
account; your harness runs it. Nothing here depends on the other fulcra-agent-*
skills; the last section shows the fleet upgrade path if you have a coord team.
Install
Three commands to a routable ledger. The engine is stdlib-only and installs on its own:
uv tool install fulcra-api # the `fulcra` CLI: auth + the Fulcra File Store (the bus)
uv tool install "git+https://github.com/ashfulcra/[email protected]#subdirectory=packages/coord-engine"
fulcra auth login # browser sign-in; an account is created on first login
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.
- 2d ago Changed 7dfd770c626d
- 3d ago Changed 43f5ddc62b91
- 11d ago First seen · 324 lines · 91 tokens per session scan A e56e8d2dddf6
fulcra-agent-atc is a skill published in the GitHub repository ashfulcra/fulcra-tools (10 stars, last pushed yesterday), licensed MIT. It adds 91 tokens to every session and 4,767 once invoked, about $0.0005 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
magic-compact
Compact this Claude Code session.
citation-intelligence
Use when the user wants to know which URLs AI engines cite for a query, whether their domain is being cited by ChatGPT/Claude/Perplexity/Gemini/Google AI Overviews/Bing, what queries their site is cited for, how citation rate changes over time, or how their citation coverage compares to competitors. Self-hosted, BYO…
proof-engine
Create formal, verifiable proofs of claims with machine-checkable reasoning. Use when asked to prove, verify, fact-check, or rigorously establish whether a claim is true or false — mathematical, empirical, or mixed. Trigger phrases: "is it really true", "can you prove", "verify this", "fact-check this", "prove it"…
join
Join the Cortex agent team — reads agent config from team directory, generates .cortex.md protocol, updates CLAUDE.local.md. Idempotent — safe to re-run as a sync.
setup
First-time Cortex setup — creates team directory, config, and chief of staff agent.
leave
Leave the Cortex agent team — removes .cortex.md, cleans CLAUDE.local.md, sets agent status to inactive in team directory.