add-skill
Install a pinned skill extension on demand (/add-skill <id>), or list core packs and extensions
Status rundown for a returning user: decided while away, needs your call, in flight, blocked, next
> /plugin marketplace add solanabr/ai-kit > /plugin install solana-ai-kit@stbr
How it fires
How this command gets triggered: by you, by Claude, or both.
/rundownContext preview
What this command does when you run it.
Status rundown for a returning user: decided while away, needs your call, in flight, blocked, next
description: "Status rundown for a returning user: decided while away, needs your call, in flight, blocked, next"
Bring the user back up to speed from facts gathered now, not from memory. Arguments, if any: `$ARGUMENTS`. They may narrow the scope (a PR, a branch, `since <date>`); `--handoff` also writes the rundown to a file for a teammate or the next session.
Five sections, each numbering its items from 1, stand in for Done / Warnings / Open in this reply; skip an empty one. Every PR, issue, commit and file they name is a clickable link: a full URL (`https://github.com/<owner>/<repo>/pull/<n>`), or a path from the project root with its line (`src/lib.rs:42`).
1. **Decided while you were away:** what landed (commits, merges, finished lanes, with links), then the calls made without the user: one-way calls and calls that went against a recommendation or an adversary first, each with its record id and how to reverse it, then a count of the rest. 2. **Needs your call:** every open question, provisional calls to ratify included, as a numbered brief in the user's language, ordered by what it blocks and restated in full, never as a pointer to an earlier message:
1. <id> — <the question in plain words>
Recommend: <answer>, because <why>. Against: <the best case against>.
Premises: <premise> [verified: <how>]; <premise> [assumed]. Reversible | One-way. Blocks: <what waits on it>.AskUserQuestion only for at most two concrete choices, mirrored in the list. A timeout or an away notice leaves the item unanswered. 3. **In flight:** running agents and workflows, open PRs with their CI and review state, and lane worktrees with unmerged commits, each with its owner. 4. **Blocked:** what cannot move and on what (failing CI, a dead agent, an unanswered call), and what is broken or unenforced: a check that did not run, a decision in force that a test or hook should enforce but whose `enforced-by` is still `none yet`. 5. **Next:** the next actions in order, the user's instructions not yet carried out (with their date), and what to type to continue, such as `/proceed`.
Write the same rundown to `.claude/context/handoff-<YYYY-MM-DD>.md`, append the output of `bash .claude/bin/decisions.sh export --handoff`, and give the path. A full install gitignores `.claude/context/`, and the record is private: whether the file leaves this machine is the user's call. No secrets, keys or tokens go in it.
Claude Code, set up to ship on Solana. Not programs alone — the whole lifecycle: validating the idea, building the program and the app around it, shipping it, and the launch and growth work that follows.
Install a pinned skill extension on demand (/add-skill <id>), or list core packs and extensions
Short for /adversarial: run every task through proposer and adversary pairs, no triage
Put a task's load-bearing claims and key decisions through a proposer and an adversary, with proof
Audit infra security: secrets, deps, CI/CD, webhooks, AI/skill files
Audit Solana program code for exploitable bugs and write a findings report
Compare per-instruction CU with the stored baseline to catch regressions