ntm

An optional adapter for managing caller-chosen agent roles and commands in persistent tmux panes, which are terminal windows that continue running. It reports whether pane work is actually progressing rather than merely showing that a command was sent.

In plain words
What is it for?
Use it when coordinating agents in tmux panes with explicit roles, commands, and write limits. It helps inspect liveness, rescue stalled work, and restart a pane only when the available evidence shows it is inactive.
Why use it?
It helps distinguish delivery of a prompt from real progress by checking new files, transcript activity, and pane state. It also preserves evidence from stalled panes before restarting them.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/boshu2/agentops/ntm
Any agent
npx skills add boshu2/agentops --skill ntm
Clone the repo
git clone --depth 1 https://github.com/boshu2/agentops

Made for: Claude Code, Codex.

Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,072 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00038 $0.01072
Opus 5 $0.00019 $0.00536
Sonnet 5 $0.00008 $0.00214
Haiku 4.5 $0.00004 $0.00107

Measured 3d ago against content hash 1aba53e6d6c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ntm 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 3d ago.

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.

images/gemini/skills/ntm/SKILL.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

NTM — optional pane adapter

NTM hosts explicit agent roles in persistent panes. It is transport, not an AgentOps lifecycle controller. The caller chooses the panes, roles, commands, write scopes, and stopping point.

Robot surfaces work because they report what the pane is doing, not what was sent to it; a dispatch layer that only proved delivery would let every dead worker look busy.

Judge pane liveness by the truth-stack, strongest first: new artifacts on disk, then transcript growth, then robot state and attention flags, then bare process existence. A successful prompt send sits below all of these and proves nothing about work.

Named failure mode — kill-the-witness: restarting a stuck pane before capturing its state, destroying the only evidence of why it stalled.

Anti-pattern: restarting an unresponsive pane as the default remedy. Corrective: rescue before restart — snapshot robot state and transcript, attempt a nudge, and restart only when the truth-stack shows no liveness at any level.

Boundary

  • Never start or probe NTM merely because it is installed.
  • Discover the live command contract with ntm --help, ntm --robot-capabilities, and ntm --robot-snapshot before unfamiliar actions.
  • Dispatch each caller-supplied command once. NTM does not select work, retry a failed command, validate a candidate, integrate changes, or decide what runs next.
  • Pane roles are descriptive. They grant no ownership, admission, Git, release, or delivery authority.
  • Concurrent writers require caller-supplied disjoint scopes and whatever isolation the repository requires. NTM does not infer safe concurrency.
  • Safety, lock, mail, and pipeline features remain substrate facts. Their state cannot change an RPI phase result or semantic verdict. The adapter cannot select AgentOps semantics, issue a binding verdict, or turn factory completion into delivery or validation proof.

One-shot dispatch

  1. Record the explicit session, pane, role, working directory, and command.
  2. Inspect robot capabilities and current snapshot.
  3. Create or select only the named pane.
  4. Send the command once.
  5. Observe robot state, attention, and transcript until the command exits or the caller's observation window ends.
  6. Return the factual result and stop.

Read the full file on GitHub · 118 lines

Changes

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.

  1. 3d ago First seen · 118 lines · 38 tokens per session scan A 1aba53e6d6c2

Subscribe to this mod's changes

ntm is a skill published in the GitHub repository boshu2/agentops (431 stars, last pushed 5d ago), licensed Apache-2.0. It adds 38 tokens to every session and 1,072 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.

Related

Other skills, from other repositories

brainstorm

Explore vague or ambitious ideas into a right-sized requirements-only plan. Use when the user wants to brainstorm, think through scope, decide what to build, or needs collaborative product framing before planning, not for a decisive verdict on whether to adopt or switch to a specific external technology, library, or…

OutlineDriven/odin-claude-plugin · 85 tokens

doc-review

Use when the user asks to review or critique a prose planning document — a plan, spec, PRD, requirements doc, or design doc.

OutlineDriven/odin-claude-plugin · 32 tokens

audit-project

Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".

OutlineDriven/odin-claude-plugin · 50 tokens

autolearn

Compound a solved problem into a durable in-repo learning doc. Use when a verified non-trivial fix lands, the user says "compound this", "document this fix", or "remember this". This is the automatic-capture entry point; for an explicitly requested one-off write-up, use compound.

OutlineDriven/odin-claude-plugin · 65 tokens

doubt-driven

Doubt-driven adversarial review. Use when correctness matters more than speed, the code is unfamiliar, stakes are high, a claim can't be checked by the type system or compiler, or verifying now is cheaper than debugging later.

OutlineDriven/odin-claude-plugin · 49 tokens

drift-detect

Use when the user says "plan drift", asks whether the roadmap, plans, or docs still match the code, or is deciding what to rebuild when restarting a stalled project. For doc-vs-code drift inside a specific diff, use sync-docs.

OutlineDriven/odin-claude-plugin · 56 tokens