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 skills/dashpay/platform/protocol-upgrade-testnpx skills add dashpay/platform --skill protocol-upgrade-testgit clone --depth 1 https://github.com/dashpay/platformWhat 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 | $0.00090 | $0.01872 |
| Opus 5 | $0.00045 | $0.00936 |
| Sonnet 5 | $0.00018 | $0.00374 |
| Haiku 4.5 | $0.00009 | $0.00187 |
Grade A, and why
protocol-upgrade-test 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 — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Protocol Upgrade Test
Review the consensus path before designing tests. Treat protocol selection, first-block state migration, and release orchestration as separate layers with separate evidence.
Authority
Default to review, which is read-only. It permits local source inspection and
test/report design only. It does not authorize ref fetching, dependency
installation, image pulls, Docker changes, network starts, config migration, or
service restarts.
Use another mode only when the user explicitly requests it:
drive: add and run deterministic Drive/ABCI tests.local-release: operate an isolated local network and release images.snapshot: inspect a supplied state artifact read-only and mutate private copies only.all: union of the explicitly requested modes.
Never broaden authority because a later step would be convenient.
Review workflow
Read references/review-checklist.md before reviewing code.
1. Resolve the comparison
For a released upgrade, require both baseline and target to be exact local
release-tag names. Resolve each tag to its peeled commit SHA and confirm the
Dashmate/package version matches the tag. Do not use a branch, HEAD, or a
mutable remote-tracking ref as evidence for code shipped in release artifacts.
If a tag is missing, stop unless the user explicitly authorizes fetching it.
Run:
.claude/skills/protocol-upgrade-test/scripts/review_transition.sh \
<baseline-tag> <target-tag> [old-protocol] [new-protocol]
For pre-release preparation only, a fetched development branch may be used as a supplemental target:
.claude/skills/protocol-upgrade-test/scripts/review_transition.sh \
--allow-non-tag-target <baseline-tag> <target-ref> \
[old-protocol] [new-protocol]
Pin and report the target SHA and label the result pre-release; it cannot
certify a release or substitute for rerunning the review against the eventual
release tag. Fetching branches or tags still requires explicit user authority.
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.
- 2d ago First seen · 201 lines · 90 tokens per session scan A a1dd0377774e
protocol-upgrade-test is a skill published in the GitHub repository dashpay/platform (92 stars, last pushed 2d ago), licensed MIT. It adds 90 tokens to every session and 1,872 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
rtc-balance
Check RustChain wallet balance, epoch info, and network status via the public RPC.
frontend-design
Create distinctive, production-grade Stellar dApp UI. Combines bold aesthetic direction with smart contract integration patterns — wallet connection flows, transaction UX, contract data displays, and agentic kit hook wiring. No generic "AI slop" aesthetics.
stellar-mcp
MCP (Model Context Protocol) tools for Stellar development. Gives Claude direct access to the Stellar blockchain, filesystem, GitHub, and Playwright through configured MCP servers. Use when querying chain data, deploying contracts, managing repos, or running e2e tests from within the agent.
dero
Query the DERO privacy blockchain and its documentation through the dero-mcp-server. Use for any DERO ecosystem question — node setup, smart contracts (DVM-BASIC), wallets, TELA apps, RPC methods, ports, privacy mechanics, transaction tracing, contract inspection.
cad-reference
Index of the Convex Architecture Documents (CADs), the normative specifications for Convex. Use to find which CAD governs a subsystem before changing protocol, encoding, consensus or economic behaviour.
cad3-encoding
CAD3 encoding format — cells, embedded vs branch references, value IDs, validity rules. Use when working on encoders, decoders, serialisation, hashing or anything that reads or writes cell encodings.