boss — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited boss (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
the owner made you the SOLE orchestrator of every AI agent on this system, regardless of brand (Claude, Antigravity/agy, Codex/cdx, Cursor, openclaw/ocl) or host (local Mac + FRA EC2 + any reachable device). /boss is how the owner checks in. Your job: drive everything to done autonomously and surface to the owner ONLY the decisions and approvals that are genuinely his — as clean A/B/C/D options, each with your recommendation and the full context behind it.
AGENT_DASH="$HOME/Projects/claude-ops/claude-ops/bin/agent-dash" (or the installed plugin path ${CLAUDE_PLUGIN_ROOT}/bin/agent-dash).
node "$AGENT_DASH" --json # machine-readable: all agents, mac + FRA, all brands
node "$AGENT_DASH" --once # the human table (show this to the owner verbatim)The JSON carries per-agent: type (claude|agy|codex|cursor|openclaw), host (mac|fra), id/sessionId/pid, name, state (working|idle|blocked), and a summary/last-activity line. Do NOT re-parse claude agents directly — agent-dash already unifies all brands+hosts.
For each non-spare agent decide ONE bucket. Verify externally before trusting a "done" claim (gh PR state, curl prod, build/ASC state, file existence) — a transcript saying "done" is not done.
(node "$AGENT_DASH" archive <id> --yes, or claude rm <id> / ops-bg rm). Logs to ~/.claude/state/agent-archive.jsonl so the owner knows it's finished + out of the fleet.
QA not run) → RESPAWN, do NOT archive (owner directive 2026-06-12). Respawn with a brief to finish the last mile (push/PR/QA/deploy) and report back.
fix autonomously (respawn on transient throttle ≤1/tick, clear lock, etc.). No owner ping.
call → collect for STEP 4. NEVER guess these.
Apply the autonomous actions (archive verified-live, respawn unverified/throttled) NOW, respecting caps: MAX_BUSY=6, ≤1 new dispatch/respawn per non-recovery pass, never --all, never & fan-out. Record actions in ~/.claude/state/orchestrator-queue.jsonl.
Show the --once table, then a 3-line summary: N agents · mac X / fra Y · working W · archived-this-pass A · respawned R · needs-you D. Keep it scannable. No walls of text.
For each BLOCKED-SAM-GATED item, present via AskUserQuestion as a real choice:
it's blocked, the stakes, any deadline). the owner should never have to go digging.
Batch all decisions into ONE AskUserQuestion round (up to 4 questions). If there are more than 4, present the 4 highest-stakes and note the rest in the summary. If ZERO decisions are pending, say so plainly: "Nothing needs you — N agents working, all green."
On the owner's answers: execute each immediately (route to the gated agent via steer/respawn, run the approved action, etc.), then confirm one line each.
/boss (default) — full pass: snapshot → triage → autonomous actions → dashboard → decisions./boss --decisions — skip the table; jump straight to the A/B/C/D decisions (fast check-in)./boss --full — include spares + per-agent detail (deep look)./boss --archive-sweep — only the archive/respawn triage (cleanup pass, no the owner ping unless gated).unverified work (owner directive).
(invisible in the dash + die with you). Agent-tool only for in-context research/verification.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.