tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
72
66%
Does it follow best practices?
Impact
80%
1.37xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./nanobot/skills/tmux/SKILL.mdQuality
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is concise and technically specific, clearly communicating the skill's unique capability of controlling tmux sessions via keystrokes and reading output. Its main weakness is the absence of an explicit 'Use when...' clause, which limits Claude's ability to know exactly when to select it. Adding trigger guidance and a few more natural user terms would strengthen it.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user needs to interact with tmux sessions, automate terminal commands, send keystrokes to running processes, or read terminal output.'
Include additional natural trigger terms users might say, such as 'terminal multiplexer', 'send-keys', 'capture-pane', 'automate terminal', or 'control running process'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'remote-control tmux sessions', 'sending keystrokes', and 'scraping pane output'. These are specific, actionable capabilities rather than vague language. | 3 / 3 |
Completeness | Clearly answers 'what' (remote-control tmux sessions by sending keystrokes and scraping pane output), but lacks an explicit 'Use when...' clause specifying when Claude should select this skill. The 'when' is only implied. | 2 / 3 |
Trigger Term Quality | Includes good terms like 'tmux', 'keystrokes', 'pane output', and 'interactive CLIs', but misses common user variations like 'terminal multiplexer', 'send keys', 'tmux send-keys', 'capture-pane', or 'screen scraping'. Users might also say 'automate terminal' or 'control terminal session'. | 2 / 3 |
Distinctiveness Conflict Risk | Very distinct niche — tmux session control via keystrokes and pane scraping is highly specific and unlikely to conflict with other skills. The combination of 'tmux', 'keystrokes', and 'scraping pane output' creates a clear, unique identity. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with excellent concrete examples and commands that are copy-paste ready. Its main weaknesses are the lack of explicit validation/error-recovery steps in workflows and the somewhat lengthy inline content (especially the orchestration section) that could benefit from being split into a separate file. The content is well-structured with clear headers but could be more concise in places.
Suggestions
Move the 'Orchestrating Coding Agents' section to a separate file (e.g., ORCHESTRATION.md) and link to it from the main skill to improve progressive disclosure and conciseness.
Add explicit validation steps after key operations — e.g., after send-keys, capture-pane to verify the command was received; after session creation, verify with list-sessions.
Remove the Windows/WSL section or reduce to a single line — platform gating in frontmatter already handles this, and Claude doesn't need the explanation.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient with minimal fluff, but the orchestrating coding agents section is quite lengthy and could be a separate file. The Windows/WSL section and some tips are somewhat unnecessary for Claude. The content is mostly lean but has room for tightening. | 2 / 3 |
Actionability | Excellent executable examples throughout — the quickstart, send-keys commands, capture-pane, wait-for-text.sh usage, cleanup commands, and orchestration examples are all copy-paste ready with concrete flags and arguments. | 3 / 3 |
Workflow Clarity | The quickstart provides a clear sequence for session creation, but there are no explicit validation checkpoints or error recovery steps. For operations like sending keys to tmux sessions (which can fail silently), there's no feedback loop for verifying commands were received correctly beyond the wait-for-text helper. | 2 / 3 |
Progressive Disclosure | Well-organized with clear section headers and references to helper scripts via `{baseDir}/scripts/`. However, the orchestrating coding agents section is quite long and inline — it would be better as a separate reference file. The content is borderline monolithic for a SKILL.md overview. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 9 / 11 Passed | |
- Repository
- HKUDS/nanobot
- Commit
04a41e3
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.