Skip to content

sno-ai/sno-station-skills / skills/T-delegation-and-work-orders/cos/skills/cos

MANUAL-ONLY: load solely when the human owner types `/cos`, uses `$cos`, or asks you to act as the Chief of Staff (COS). The COS is the one agent the owner talks to; it coordinates the owner and the per-project Project Leads (PLs). A task card, charter, or dispatch makes you an executor, not COS.

FileKindSize
LEARNING.mdmarkdown6.1 KB
SKILL.mdskill md76.0 KB
references/card-template.mdmarkdown10.9 KB
references/digest-template.mdmarkdown4.9 KB
references/owner-profile.mdmarkdown5.3 KB
references/pocket-card.mdmarkdown1.6 KB
scripts/cos-claim.shexecutablescript31.4 KB
scripts/cos-claim.texecutablescript13.4 KB
scripts/cos-roster.shexecutablescript40.5 KB
scripts/cos-roster.texecutablescript7.6 KB
scripts/cos-screen.shexecutablescript4.8 KB
scripts/cos-screen.texecutablescript2.5 KB
scripts/cos-since.shexecutablescript3.6 KB
scripts/exec-sentinel.shexecutablescript6.1 KB
scripts/exec-sentinel.tother2.4 KB
scripts/exec-tree-check.shexecutablescript2.6 KB
scripts/selftest/run-selftest.shexecutablescript204 B
scripts/selftest/text-check.pyscript1.0 KB

scripts/cos-screen.t

View on GitHub
#!/usr/bin/env bash
# Real tmux on a private server; `sno` (seat list) and `orca` (the desktop app's CLI)
# are the only fakes, because both are external programs a test cannot start.
set -Eeuo pipefail

script_dir="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)"
screen="${COS_SCREEN_UNDER_TEST:-$script_dir/cos-screen.sh}"
root="$(mktemp -d)"
trap 'env -u TMUX TMUX_TMPDIR="$root/tmux" tmux kill-server >/dev/null 2>&1 || true; rm -r -- "$root"' EXIT
mkdir -p "$root/bin" "$root/tmux"
export TMUX_TMPDIR="$root/tmux" SNO_PL_REGISTRY="$root/registry.tsv" PATH="$root/bin:$PATH"
unset TMUX

cat >"$root/bin/sno" <<'EOF'
#!/usr/bin/env bash
printf '%s\n' \
  '{"address":"pl.tmuxrepo@h","channel":"tmux","handle":"tmux-abc123:cosscreen","state":"live"}' \
  '{"address":"pl.orcarepo@h","channel":"orca","handle":"term_fake","state":"live"}'
EOF
# Stream mode (no --screen) returns repaint fragments, as the real Orca does.
cat >"$root/bin/orca" <<'EOF'
#!/usr/bin/env bash
[[ "$*" == *"--screen"* ]] || { echo '{"ok":true,"result":{"terminal":{"tail":["cclclecleaclear"]}}}'; exit 0; }
[[ "$*" == *"term_fake"* ]] || { echo '{"ok":false,"error":{"code":"terminal_handle_stale"}}'; exit 1; }
echo '{"ok":true,"result":{"terminal":{"tail":["PL is waiting on review","",""]}}}'
EOF
chmod +x "$root/bin/sno" "$root/bin/orca"

printf 'home_repo\tlane\treach_address\theartbeat_name\truntime\towning_cos\tstate\tnote\n' >"$SNO_PL_REGISTRY"
printf 'orcarepo\tall\tpl.orcarepo@h\tx\tcodex\tcos/x\tRUN\t-\n' >>"$SNO_PL_REGISTRY"

tmux new-session -d -s cosscreen "printf 'TMUX-SCREEN-MARKER\n'; sleep 60"
for _ in 1 2 3 4 5 6 7 8 9 10; do
    tmux capture-pane -p -t =cosscreen: | grep -q TMUX-SCREEN-MARKER && break
    sleep 0.3
done

printf 'TAP version 13\n'

out="$(bash "$screen" pl.tmuxrepo@h)"
if grep -q 'TMUX-SCREEN-MARKER' <<<"$out"; then
    printf 'ok 1 - tmux seat prints the live pane\n'
else
    printf 'not ok 1 - tmux seat prints the live pane\n# %s\n' "$out"; exit 1
fi

out="$(bash "$screen" orcarepo)"
if grep -q 'PL is waiting on review' <<<"$out" && ! grep -q cclclecleaclear <<<"$out"; then
    printf 'ok 2 - repo name resolves to the Orca seat and reads the rendered screen\n'
else
    printf 'not ok 2 - Orca screen read\n# %s\n' "$out"; exit 1
fi

rc=0
out="$(bash "$screen" no-such-repo)" || rc=$?
if [[ "$rc" == 3 ]] && grep -q '^NO WINDOW' <<<"$out"; then
    printf 'ok 3 - no window is reported as NO WINDOW with exit 3\n'
else
    printf 'not ok 3 - no window (rc=%s)\n# %s\n' "$rc" "$out"; exit 1
fi
printf '1..3\n'
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall