Jolt is a zero-knowledge virtual machine for RISC-V that enables programs to be executed and cryptographically proven. It is for developers and contributors working with zero-knowledge proofs and virtual machines, while the catalogue entries provide instructions and skills for using or developing Jolt.
Borrowing it
Nothing to install: this file belongs to a16z/jolt. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/a16z/jolt/main/.claude/skills/analyze-spec/SKILL.mdgit clone --depth 1 https://github.com/a16z/joltWrote 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/a16z/jolt/analyze-spec)<a href="https://agentmods.dev/skills/a16z/jolt/analyze-spec"><img src="https://agentmods.dev/badge/skills/a16z/jolt/analyze-spec/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/a16z/jolt/analyze-spec"><img src="https://agentmods.dev/badge/skills/a16z/jolt/analyze-spec.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 | $0.00018 | $0.01254 |
| Opus 5.5 | $0.00007 | $0.00502 |
| Sonnet 5.5 | $0.00004 | $0.00251 |
| Haiku 4.5 | $0.00002 | $0.00125 |
Grade A, and why
analyze-spec 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 1mo 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This skill operates in two modes:
- Local mode (invoked via
/analyze-specin Claude Code): Full interactive Socratic interview — one question at a time, iterative refinement with the spec author. - Remote mode (triggered externally via the
claude-spec-review-requestlabel): Single-pass analysis — all questions posted at once as a single PR comment. Reads prior PR comments as context to account for already-answered questions.
Adapted from the Ouroboros-inspired deep interview methodology — specification quality is the primary bottleneck in AI-assisted development.
<Execution_Policy>
- Gather codebase facts via
exploreagent BEFORE asking about them - Cite repo evidence (file path, symbol, or pattern) instead of asking the spec author to rediscover it
- Score ambiguity transparently
- Do not approve until ambiguity ≤ threshold (default 0.2)
- Allow early approval with a clear warning if ambiguity is still high </Execution_Policy>
<Mode_Detection> Detect which mode to use:
- Remote mode: Running in a remote Claude instance triggered by the
claude-spec-review-requestlabel. Indicators: environment is non-interactive (no TTY), or running inside a GitHub PR context. - Local mode: Running interactively in a terminal via
/analyze-spec.
When in doubt, default to local mode (interactive). </Mode_Detection>
Phase 1: Initialize
- Locate the spec:
- If a path is provided in
{{ARGUMENTS}}, use that directly. - Otherwise, detect the PR number:
- Run
gh pr view --json number --jq .numberto get the current branch's PR number. - If that fails, list specs:
ls specs/and ask the user which one.
- Run
- Look for
specs/<PR#>-*.mdmatching the PR number. This is the spec for this PR. - If no match, fall back to finding any
specs/*.mdfile that is NOTTEMPLATE.md. - If multiple specs match, prefer the one matching the PR number. If still ambiguous, ask the user.
- If a path is provided in
- Read the spec thoroughly — understand all sections (Summary, Intent, Evaluation, Design, Execution).
- Read
jolt-eval/README.mdso you understand the invariant/objective framework for scoring Success Criteria and generating questions. - Explore the codebase: Run
exploreagent to map codebase areas relevant to the spec's intent. - Read prior context (remote mode): Read all existing PR comments via
gh pr view --json commentsto identify questions already asked and answers already given. Account for these when scoring — don't re-ask answered questions.
What ships with it
2 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.
- 1mo ago Changed 04928489117d
- 2mo ago First seen · 104 lines · 18 tokens per session scan A ebc8f7083970
analyze-spec is a skill published in the GitHub repository a16z/jolt (1,059 stars, last pushed today), licensed Apache-2.0. It adds 18 tokens to every session and 1,254 once invoked, about $0.0001 per session on Opus 5.5. 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
implementing-zero-knowledge-proof-for-authentication
Zero-Knowledge Proofs (ZKPs) allow a prover to demonstrate knowledge of a secret (such as a password or private key) without revealing the secret itself. This skill implements the Schnorr identificati.
zero-knowledge-proofs
Use this skill when the user says 'ZK circuit', 'Circom', 'zk-SNARK', 'Groth16', 'PLONK', 'trusted setup', 'ptau', 'privacy-preserving proof'. Covers writing arithmetic circuits (Circom), choosing a proving system, and the trusted-setup lifecycle. Do NOT use for: full ZK domain reference (use…
zkp-toolkit
Zero-Knowledge Proof utilities - proof generation, verification, selective disclosure, ZK-Rollup batching, and benchmarking across ZK systems.
configuring-hsm-for-key-storage
Hardware Security Modules (HSMs) are tamper-resistant physical devices that safeguard cryptographic keys and perform cryptographic operations in a hardened environment. Keys stored in an HSM never lea.
implementing-aes-encryption-for-data-at-rest
AES (Advanced Encryption Standard) is a symmetric block cipher standardized by NIST (FIPS 197) used to protect classified and sensitive data. This skill covers implementing AES-256 encryption in GCM m.
configuring-certificate-authority-with-openssl
A Certificate Authority (CA) is the trust anchor in a PKI hierarchy, responsible for issuing, signing, and revoking digital certificates. This skill covers building a two-tier CA hierarchy (Root CA +.