CtrlK
BlogDocsLog inGet started
Tessl Logo

delegate-to-ide

Delegate a task to another AI IDE's CLI (codex / claude / opencode / cursor-agent) through an isolated-context subagent. Use when the user says delegate to <ide>, have <ide> do <task>, or offload to <ide>. For one-shot relay use `ai-ide-runner`.

68

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

75%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.

A well-structured, actionable routing skill with a clear workflow and an explicit validation checkpoint. The main improvement areas are trimming repeated 'do not run inline' restatements and converting the per-host invocation guidance into literal code blocks.

Suggestions

Consolidate the 'do not run the child CLI inline from this parent session' rule into one authoritative location (e.g. the Output contract) and reference it, rather than restating it across ~5 sections.

Render the per-host subagent invocations as literal copy-pasteable code blocks (e.g. a full Agent/Task tool call example) instead of prose descriptions, to push actionability toward 5.

If the Codex/Cursor installation-detection and limitation notes grow, move them into a references/ file to keep the overview lean and support progressive disclosure.

DimensionReasoningScore

Conciseness

Dense and assumes Claude's competence (no preamble about what an IDE/CLI is); the 'do not run the child CLI inline' rule is restated across ~5 sections, which is mild over-emphasis that could be trimmed, keeping it just below a 5.

4 / 5

Actionability

Gives concrete per-host invocation syntax (subagent_type=worker, @worker, spawn_agent with fork_turns="none" + wait_agent), a detection command (`ls "${CODEX_HOME:-$HOME/.codex}/agents/"`), and a worker block template; not a 5 because the invocations are prose-described rather than literal copy-paste code blocks.

4 / 5

Workflow Clarity

A clear 3-step sequence (parse intent → invoke worker → relay reply) with a validation checkpoint (the agents-dir ls check) and explicit error routing ('empty listing means the role is not installed — say so'); not a 5 because the validation is a single light check rather than a full validate-fix-retry loop.

4 / 5

Progressive Disclosure

Self-contained with clear section headers (When to use, Workflow, Output contract, Scope boundaries) and no bundle files; at ~100 lines it exceeds the 50-line simple-skill threshold, so it sits at 4 rather than 5 despite good organization.

4 / 5

Total

16

/

20

Passed

Description

87%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.

A strong, specific description that clearly states both capability and trigger conditions and actively disambiguates from a sibling skill. The only mild gap is breadth of action verbs, which keeps specificity and trigger quality at 4 rather than 5.

DimensionReasoningScore

Specificity

Names the domain and a concrete mechanism ('through an isolated-context subagent') plus the four target CLIs (codex / claude / opencode / cursor-agent); not a 5 because it lists essentially one action verb rather than multiple distinct capabilities.

4 / 5

Completeness

Explicitly answers what ('Delegate a task to another AI IDE's CLI ... through an isolated-context subagent') and when ('Use when the user says delegate to <ide>, have <ide> do <task>, or offload to <ide>') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Three natural user phrasings are given ('delegate to <ide>', 'have <ide> do <task>', 'offload to <ide>') with synonyms; not a 5 because file-extension-style tokens do not apply here and a few natural variations are absent.

4 / 5

Distinctiveness Conflict Risk

Clear niche (delegated task execution via an isolated-context subagent) and it explicitly disambiguates from the sibling skill ('For one-shot relay use `ai-ide-runner`'), minimizing conflict risk.

5 / 5

Total

18

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
korchasa/flowai-plugins
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.