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/untrivial-ai/agent-orchestrator/bug-triagenpx skills add Untrivial-ai/agent-orchestrator --skill bug-triagegit clone --depth 1 https://github.com/Untrivial-ai/agent-orchestratorWhat 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 | $0.00033 | $0.04611 |
| Opus 5 | $0.00016 | $0.02305 |
| Sonnet 5 | $0.00007 | $0.00922 |
| Haiku 4.5 | $0.00003 | $0.00461 |
Grade A, and why
bug-triage 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 yesterday.
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 — 422 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bug Triage Skill
Triage bugs into well-structured GitHub issues on the Agent Orchestrator repo.
Agent Orchestrator is Go + Electron. The backend is a Go daemon (
backend/) exposing a loopback HTTP API on127.0.0.1:3001; the frontend is an Electron + React supervisor (frontend/). There is no pm2/tmux/Node runtime here — the daemon owns lifecycle and sessions run under the Zellij runtime adapter. Triage against this stack, not the old TypeScript agent-orchestrator.
⚠️ Which ao are you running?
Agent Orchestrator ships no ao on your PATH. A bare ao very likely resolves to a
different AO install — e.g. an old npm build at ~/.nvm/.../bin/ao that talks
to port :3000. Triaging with the wrong binary produces bugs that don't exist
in Agent Orchestrator (and miss ones that do).
Before any diagnostics:
which -a ao # see every ao on PATH — expect surprises
ao status 2>/dev/null # if this shows port 3000, it is NOT Agent Orchestrator
Use an Agent Orchestrator binary explicitly:
# Option A — build from this repo (preferred during triage)
cd backend && go build -o /tmp/ao ./cmd/ao
/tmp/ao status # must report port: 3001
# Option B — the packaged app's bundled daemon
"/Applications/Agent Orchestrator.app/Contents/Resources/daemon/ao" status
Confirm ao status reports port: 3001 before trusting any output. Throughout
this skill, ao means your verified Agent Orchestrator binary (/tmp/ao or the bundled
one), never a bare PATH lookup.
Note: spawned sessions get a PATH pin so the session's
aoresolves to the daemon's own executable (seehookPATHinbackend/internal/session_manager/manager.go). That pin only applies inside sessions — your interactive shell is still on its own PATH, so pin it yourself.
1. Pre-flight
- Pull latest code:
git pull origin main. Stale code = bad triage. - Target repo: Always file on
Untrivial-ai/agent-orchestrator(the current product repo, not a fork). Agent Orchestrator is the product, not a thin fork of upstream. - Verify your binary: confirm
ao statusshows port 3001 (see warning above). - Record source: chat URL, reporter name, attachments.
What ships with it
1 file 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.
- yesterday First seen · 422 lines · 33 tokens per session scan A 2c6df9861f08
bug-triage is a skill published in the GitHub repository Untrivial-ai/agent-orchestrator (10,763 stars, last pushed today), licensed Apache-2.0. It adds 33 tokens to every session and 4,611 once invoked, about $0.0002 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
bernstein-run
Run a verified multi-agent goal with Bernstein. Use when a task is too large for a single agent session: Bernstein decomposes the goal into tasks, spawns CLI coding agents in parallel git worktrees, verifies their output, and merges results. Also use to check run status, costs, and to verify a finished run against its…
bernstein-plan
Create and manage multi-step execution plans in Bernstein. Plans decompose complex goals into stages with dependencies. Use when the user wants to plan a complex feature, break down a large task, or review an execution plan before agents start working.
bernstein-agents
Manage Bernstein agents - list active agents, inspect their output, kill stalled agents, or stream live logs. Use when the user asks about agents, wants to see what an agent is doing, or needs to kill one.
bernstein-approve
Review and approve/reject pending tasks or plans in Bernstein. Use when the user asks about approvals, wants to review agent work, or needs to approve/reject a plan before execution begins.
bernstein-quality
Show quality metrics for Bernstein runs - success rates per model, lint/test pass rates, completion time distributions. Use when the user asks about quality, reliability, which model performs best, or pass rates.
manager
Planning - decompose goals, create tasks via task server.