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 agents/verified-zkevm/evm-asm/sail-phase2-bootstrapgit clone --depth 1 https://github.com/Verified-zkEVM/evm-asmWrote 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/agents/verified-zkevm/evm-asm/sail-phase2-bootstrap)<a href="https://agentmods.dev/agents/verified-zkevm/evm-asm/sail-phase2-bootstrap"><img src="https://agentmods.dev/badge/agents/verified-zkevm/evm-asm/sail-phase2-bootstrap.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 | $0.00000 | $0.02520 |
| Opus 5 | $0.00000 | $0.01260 |
| Sonnet 5 | $0.00000 | $0.00504 |
| Haiku 4.5 | $0.00000 | $0.00252 |
Grade A, and why
sail-phase2-bootstrap 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 6d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sail-zkVM integration — Phase 2 bootstrap (foundation migration)
Historical bootstrap record. Current maintenance moved to
Verified-zkEVM/riscv-zkvm.
For: the next session. Goal of P2: vendor the (now-proven) current-upstream
scoped RV64IM Lean model into evm-asm, build it in-project, drop the moving
dhsorens fork, and re-establish the 51 *_sail_equiv lemmas against it.
Status (2026-06-25): steps 1–4 DONE — the core P2 migration is complete and
green. dhsorens is gone from lakefile.toml/lake-manifest.json; the project
builds against the vendored model (require out from "vendor/sail-riscv-zkvm-lean").
Full lake build EvmAsm = 2984/2984, exit 0. All 51 *_sail_equiv lemmas green;
axioms = {propext, Classical.choice, Quot.sound} only (no custom axioms, no sorryAx);
no forbidden tactic in vendored or proof code. The LeanRV64D → Out rename was 100%
mechanical — the feared bool_to_bit → bool_to_bits rename was a PHANTOM (both old and
new model use bool_to_bit; the doc below had it backwards). Residual P2 items 5–7
(config revalidate, scoping write-up, pin gate) are tracked below and in
docs/agents/sail-phase3-bootstrap.md; none block downstream proof work.
What proved out (the two real risks, both retired):
- lean-sail v4 + mathlib-
mastercoexist in the real evm-asm build (StateRel + 2984-job full build). This was the one unverified step-1 risk. - The 51 lemmas re-point with zero genuine logic renames — pure
LeanRV64D → Out.
Read first:
sail-import/PROVENANCE.toml [target](the proven pins +vendor_path
model_sha256). Thenscripts/regen-sail-model.sh(the validated regen recipe) anddocs/agents/sail-regen-spike.md(the GO report) only if you need to regenerate.
Proven foundation (DONE — don't re-litigate)
- Toolchain stack: Sail 0.20.2 (opam switch
sail5, OCaml 5.4.1 — 4.14.2 stack-overflows, sail#1674) + sail-riscv release tag2026-06-22-b5a2182+ external lean-sail v4 (79b4d08) on lean4 v4.30.0-rc1 (NO bump). z3 4.15.3 at the nix path inregen-sail-model.sh(machine-specific — de-nix before CI). - Generation works (step ✔):
scripts/regen-sail-model.sh --run <dir>→ 113 files, scopedmain I_insts M_insts, decodebv_decide-free. - Matched pair builds (step ✔): that model + lean-sail v4 →
lake build84/84, exit 0. - VENDORED + re-proven from git (step 2, commit
a2dedf448):vendor/sail-riscv-zkvm-lean/(113.lean, packageout, libOut). Itslakefile.tomlgit-pins lean-sail @79b4d08(tag v4);lean-toolchain= v4.30.0-rc1. A cleanlake buildfrom the committed dir — fetching lean-sail fresh from the pinned rev — gives 84/84, exit 0.model_sha256recorded in PROVENANCE. Package/lib names kept verbatim from the proven artifact (renaming the lib would rewrite every internalimport Out.*).
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.
- 6d ago First seen · 130 lines · 0 tokens per session scan A 25ad3df9cdfd
sail-phase2-bootstrap is an agent published in the GitHub repository Verified-zkEVM/evm-asm (55 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,520 tokens. 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.