coding-agents-farm

To orchestrate parallel coding-agent farms (Claude, Codex, Copilot, Gemini, etc.) on isolated git worktrees.

Quality

58%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./instructions/r2/core/skills/coding-agents-farm/SKILL.md

Quality

Content

63%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 skill body is highly actionable with a well-sequenced, validation-backed workflow for a genuinely dangerous batch operation, supported by justified reference tables. Conciseness and progressive disclosure are the weaker dimensions due to duplicated lines and large inlined tables with no reference files.

Suggestions

Remove the duplicated OpenCode/Copilot/Goose model-selection lines (the second occurrence of each) to tighten conciseness.

Move the CLI auto-approve flag tables and model-selection guidance into a reference file (e.g., references/cli-reference.md) and link to it from SKILL.md to improve progressive disclosure.

Tighten validation checkpoints with exact commands (e.g., a concrete test command in the Collect/Merge validate step) to push workflow clarity higher.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and the CLI/model tables are justified by the explicit 'your knowledge is stale' note, but the OpenCode/Copilot/Goose model-selection lines are duplicated verbatim and a few asides could be trimmed.	3 / 5
Actionability	Provides concrete, mostly copy-paste-ready commands per CLI (claude -p ..., codex exec ..., git worktree add ...) with exact flags; minor gaps remain where PROMPT/MODEL placeholders must be filled in.	4 / 5
Workflow Clarity	The 8-step process (Plan through Report) is clearly sequenced with validation checkpoints (overlap check, user approval, clean-worktree verify, pre-flight test, test-pass validation in merge) and intervention feedback loops, though some checkpoints are described loosely without exact commands.	4 / 5
Progressive Disclosure	Sections are well organized with XML-style headers and external resource links, but no bundle/reference files exist and the large CLI-flag and model-selection tables that could live in separate files are fully inlined in a ~175-line SKILL.md.	3 / 5
	Total	14 / 20 Passed

Description

53%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 clearly states what the skill does and targets a distinct niche, but it lacks an explicit 'when to use' trigger clause and only names a couple of concrete actions. Natural-language trigger coverage is moderate.

Suggestions

Add a 'Use when ...' clause with natural trigger phrases such as 'Use when running multiple coding agents in parallel or cross-validating work across Claude, Codex, Copilot, and Gemini.'

List more concrete actions (e.g., 'decompose features, launch agents on isolated git worktrees, monitor and merge results') to raise specificity.

Include common synonyms users would say ('run agents in parallel', 'concurrent coding agents', 'multi-agent execution') to improve trigger term quality.

Dimension	Reasoning	Score
Specificity	Names the domain ('parallel coding-agent farms') and concrete actions ('orchestrate ... on isolated git worktrees') plus the target CLIs, but stops at 1-2 actions without comprehensive coverage.	3 / 5
Completeness	The 'what' is clear (orchestrate parallel coding-agent farms on isolated git worktrees) but there is no 'Use when...' or equivalent explicit trigger guidance, which caps completeness at 3.	3 / 5
Trigger Term Quality	Relevant keywords appear (Claude, Codex, Copilot, Gemini, git worktrees) but common natural phrasings like 'run multiple agents in parallel' or 'concurrent coding agents' are absent and 'farms' is jargon.	3 / 5
Distinctiveness Conflict Risk	The parallel coding-agent orchestration on isolated git worktrees niche is fairly distinct with minimal overlap risk, though 'etc.' slightly broadens it.	4 / 5
	Total	13 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: griddynamics/rosetta
Path: instructions/r2/core/skills/coding-agents-farm/SKILL.md
Commit: 044d86c

Reviewed: 21 hours ago

Table of Contents

Discovery Implementation Validation

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.