CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-rtk

Install rtk token-compression proxy into agent containers. Routes Bash tool calls through rtk for 60–90% token savings on dev commands (git, cargo, pytest, docker, kubectl, etc.).

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exemplary procedural skill: fully executable commands in a clearly sequenced five-step workflow with verification checkpoints, idempotency handling, and a troubleshooting section with recovery loops. The only minor cost is a few sentences of design rationale that could be trimmed for token efficiency.

DimensionReasoningScore

Conciseness

The body is command-first and lean — no explanations of concepts Claude already knows — but minor prose could be trimmed, e.g. the "Integration tests" design-rationale paragraph and the sentence explaining why the Verify step exists ("so a missing or non-executable mount surfaces immediately rather than as a silent hook failure"). Efficient with minor over-explanation — anchor 4, not 5.

4 / 5

Actionability

Every step is copy-paste-ready: the curl install one-liner, the `ncl groups config add-mount` invocation with flags, the idempotent `jq` hook-injection command, `ncl groups restart`, `docker exec ... rtk --version`, and `rtk gain`, plus executable troubleshooting commands. Fully executable and covering the common cases — anchor 5.

5 / 5

Workflow Clarity

Steps 1–5 are explicitly sequenced with verification checkpoints after install (`rtk --version`), after config (`ncl groups config get`), and after restart (in-container `rtk --version`), and Troubleshooting provides feedback loops (e.g. hook missing → re-run Step 4; command not found → check mount and restart). The one destructive-ish operation, the settings.json hook write, is made idempotent and validated. Clear sequence with explicit validation — anchor 5.

5 / 5

Progressive Disclosure

There are no bundle files (references/, scripts/, assets/ are absent), and this single-purpose install skill keeps all of its short, well-sectioned content (setup summary, steps, verify, troubleshooting) inline where it belongs — nothing warrants a separate file. Clear overview ("What this sets up") with well-organized sections and easy navigation.

5 / 5

Total

19

/

20

Passed

Description

66%Weight 40%Scale 1-5

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 specific, quantified, and third-person, with good natural trigger terms around rtk and token savings. Its main gap is the absence of any explicit "Use when..." trigger guidance, which caps completeness and leaves triggering to keyword overlap alone.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user asks to install or set up rtk, reduce token usage of dev commands, or compress agent command output."

Add one or two natural synonyms users would say, such as "cut token costs" or "reduce context usage", to strengthen trigger-term coverage.

Consider scoping the dev-command list (e.g. "dev CLI output") to reduce overlap with skills about git, pytest, or docker themselves.

DimensionReasoningScore

Specificity

"Install rtk token-compression proxy into agent containers" and "Routes Bash tool calls through rtk for 60–90% token savings on dev commands (git, cargo, pytest, docker, kubectl, etc.)" name concrete actions with a quantified outcome and specific commands. This matches anchor 4 (several specific actions, minor gaps) — not 5, since only two actions are described and coverage is narrower than the comprehensive anchor, and not 3, since the actions are specific and enumerated rather than generic.

4 / 5

Completeness

The "what" is clear (install rtk proxy into agent containers, route Bash calls through it for token savings), but there is no "Use when..." clause or equivalent explicit trigger guidance. Per the judging guidelines, a missing 'Use when' clause caps completeness at 3.

3 / 5

Trigger Term Quality

Includes natural terms users would say: "rtk", "token savings", "Bash tool calls", and the command names "git, cargo, pytest, docker, kubectl". Good keyword coverage, but a few natural trigger phrasings (e.g. "reduce token usage", "cut context costs") are missing, so it sits between anchors 3 and 5 at 4.

4 / 5

Distinctiveness Conflict Risk

"rtk token-compression proxy" establishes a clear, distinct niche with minimal conflict risk, but the broad dev-command list (git, cargo, pytest, docker, kubectl) creates minor overlap risk with skills about those individual tools. Mostly distinct with minor overlap — anchor 4 rather than 5.

4 / 5

Total

15

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

nanocoai/nanoclaw

Path

.claude/skills/add-rtk/SKILL.md

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.