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 gear-foundation/vara-skills --skill gear-builtin-actorsgit clone --depth 1 https://github.com/gear-foundation/vara-skillsWrote 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/gear-foundation/vara-skills/gear-builtin-actors)<a href="https://agentmods.dev/skills/gear-foundation/vara-skills/gear-builtin-actors"><img src="https://agentmods.dev/badge/skills/gear-foundation/vara-skills/gear-builtin-actors/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/gear-foundation/vara-skills/gear-builtin-actors"><img src="https://agentmods.dev/badge/skills/gear-foundation/vara-skills/gear-builtin-actors.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00097 | $0.01084 |
| Opus 5 | $0.00048 | $0.00542 |
| Sonnet 5 | $0.00019 | $0.00217 |
| Haiku 4.5 | $0.00010 | $0.00108 |
Grade A, and why
gear-builtin-actors 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 12d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gear Builtin Actors
Goal
Treat builtin actors as a distinct call target: runtime-provided pseudo-programs at hardcoded ActorIds, reached via the normal msg::send_for_reply family but with request and reply types defined in the gbuiltin-* helper crates, not in a Sails IDL.
Inputs
../../references/gear-builtin-actors.md— actor catalog, ActorId derivation, per-actor request/response shapes, calling pattern../../references/gear-messaging-and-replies.md— reply semantics forsend_for_reply../../references/gear-gas-reservations-and-waitlist.md— gas budgeting and waitlist interaction when awaiting builtin replies../../references/gear-gstd-api-and-syscalls.md— the underlyinggstd::msgAPI family
Route Here When
- a feature needs staking, nominating, proxy management, BLS12-381 pairings, or ETH bridging from inside a Sails program
- a
send_for_replytargets a hardcoded non-programActorIdand the reply does not decode as a Sails route - a reply bytes stream is expected to decode as a
gbuiltin_*::Responserather than a Sails-routed payload - a builder is unsure whether to call a builtin directly from a service handler or wrap it behind a dedicated broker service
- a gtest or local-smoke run fails at the builtin call site with gas or ED errors
- a runtime upgrade changes the set of registered builtins and existing hardcoded
ActorIdconstants need revalidation
Working Model
- Confirm the actor is registered in the target runtime. For Vara, the authoritative tuple is
BuiltinActorsingear/vara/runtime/vara/src/lib.rs(IDs 1 = BLS12-381, 2 = staking, 3 = ETH bridge, 4 = proxy). IDs are runtime-version-bound. - Pick the correct
gbuiltin-*helper crate:gbuiltin-staking,gbuiltin-proxy,gbuiltin-bls381,gbuiltin-eth-bridge. Import itsRequest(andResponsewhere defined). - Derive the
ActorIdfrom the builtin ID viahash((b"built/in", id).encode()), or take the stable hex from the reference. Store as aconst ActorIdin the module that owns the call. - Call from an async handler:
msg::send_bytes_for_reply(ACTOR_ID, &payload[..], value, reply_deposit)?.await. Value and reply deposit follow Substrate ED rules for the builtin's underlying pallet. - Classify the reply: success reply decodes as
gbuiltin_*::Response(where one is defined) or is empty for fire-and-forget variants; an error reply surfaces as the.awaitErr arm and carries aBuiltinActorErrorpayload. - Budget gas against the builtin's declared
max_gas()weight; block allowance failures manifest as error replies, not panics. - For first-time interactions with a new builtin address, confirm ED is minted (runtime migrations handle this for shipped builtins — only relevant on fresh chains or new actors).
- When a service wraps a builtin (broker pattern), keep the wrapping service responsible for idempotency, reply routing, and state reconciliation — the builtin itself has no memory, no
read_state, and no code.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 48 lines · 97 tokens per session scan A 7aa5c7405cd9
gear-builtin-actors is a skill published in the GitHub repository gear-foundation/vara-skills (17 stars, last pushed 2mo ago), licensed MIT. It adds 97 tokens to every session and 1,084 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 2026-08-30.
Other skills, from other repositories
signature-replay
Signature replay attacks — missing nonces, missing chain ID, ecrecover zero address, signature malleability, cross-chain replay.
official-sui-skills
Pointer to the official Mysten Labs skills for building on Sui — language fundamentals, object model, PTBs, SDKs, publishing, upgrades, frontend integration, accessing on-chain data. Maintained upstream at github.com/MystenLabs/skills; pinned to the same ref the audit catalog derives from (see…
bridge
Cross-chain token transfers using Wormhole and CCTP.
onchainkit
Build onchain apps with Coinbase's OnchainKit React components - wallets, swaps, NFTs, payments.
marginfi
MarginFi — Solana lending and borrowing.
crypto-market-rank
Crypto market rankings and leaderboards. Query trending tokens, top searched tokens, Binance Alpha tokens, tokenized stocks, social hype sentiment ranks, smart money inflow token rankings, top meme token rankings from Pulse launchpad, and top trader PnL leaderboards. Use this skill when users ask about token rankings…