claude-kickoff

Stand up Mission Control — the live, two-way ops board for a project (SSE-live, phone-friendly, token-gated, localhost-only) with the encrypted Secrets channel built in. The coordinator writes state via mc-update.py and the operator sees it + checks items off + hands over secrets, from their phone. Use to start MC for a project, or when the operator asks "where do I see what's happening / how do I give you a key".

Stand up Mission Control — the live, two-way ops board for a project (SSE-live, phone-friendly, token-gated, localhost-only) with the encrypted Secrets channel built in. The coordinator writes state via mc-update.py and the operator sees it + checks items off + hands over secrets, from their phone. Use to start MC for a project, or when the operator asks "where do I see what's happening / how do I give you a key".

Apply this orchestration system to an EXISTING (brownfield) repo. Read the codebase, draft a CLAUDE.md + specialist subagents that match its real domains, add a tracker + memory — additively, without touching its source. Use when the human wants to run the coordinator pattern on a repo they already have.

Report a semantic beat to Mission Control — a decision reached, a milestone hit, or a completion-with-artifact (what shipped + its link). Use when you agree on an approach with the operator, hit a real milestone, get blocked/unblocked, or finish a workstream with something to show — the "why" and the artifact the automatic lifecycle hook can't see. Complements (does not replace) the spawned/working/done spine the plugin's lifecycle hook streams for free.

Take a greenfield idea to a built, tested, running first slice. Propose a tech stack (don't assume one), scaffold it FRESH with the real CLIs, wire one proof test, run it. Use when the human asks to spin up / start / bootstrap a new project or service.

Show a running result on the human's phone — or share a PUBLIC link a friend can open. Start the project's service(s), expose them over Tailscale (serve = your devices only; funnel = public, human-gated), and report ONE link. Two tiers: a single service (tailscale serve/funnel) or ≥2 services sharing an origin (the box-ingress Caddy). Use when the human wants to SEE the thing running, or SHARE it, not just read that tests passed.

Guided one-time setup of the CONTROL PLANE — wire Telegram (steer from your pocket) + Tailscale (preview links + the hosted-worker bridge), checking and verifying each step. Use when a new user wants to set up / onboard / configure the control plane, wire Telegram or Tailscale, or asks "how do I steer this from my phone."

The adversarial review turned on the CREW itself — hold each agent charter, skill, CLAUDE.md principle, and memory against RECENT OUTCOMES, find where the crew config would let a known failure recur, and close the gap. Auto-apply the ungated fixes (a stale/contradictory memory); stage the gated ones (charter/CLAUDE.md edits) as a one-tap turnkey. Fire at natural boundaries + on demand. Velocity-first triage by default.

Manage the project's harness plugins / MCP servers — the AGENT picks them by what the project needs (the operator never has to). Detect the stack + domain → install the relevant capability plugins; add one on demand when a need surfaces; prune unused ones (minimal surface = security). Handles the two gates: secret-bearing plugins need the human's key via a secure channel; keep the surface minimal.

The agent CLOSES the structural issues a non-technical operator can't — runs the scanners + reads the gate/review status, fixes the real footguns (and leaked secrets) behind the review gate, and checkpoints. Fire at session-start as a health check, and on demand. Velocity-first: session-start is a quick triage, not a refactor marathon.

Adversarial review of a diff run on LOCAL compute — fan out independent reviewers briefed to BREAK the change, verify each finding against the code, fix the confirmed ones, light re-review on the fix. Sized to the diff (trivial → skip; substantial / security / money / irreversible → run). Use before a substantial commit.

Run the generic footgun scanners (secrets + structural) over the repo and return a ranked findings list. Language-agnostic, no deps. Use at session-start (as part of harden), before a checkpoint, and on demand to answer "is anything unsafe or fragile in here?".