Name: agenthub
Rating: 65.60000000000001 (1 reviews)
Author: alirezarezvani

agenthub

Multi-agent collaboration plugin that spawns N parallel subagents competing on the same task via git worktree isolation. Agents work independently, results are evaluated by metric or LLM judge, and the best branch is merged. Use when: user wants multiple approaches tried in parallel — code optimization, content variation, research exploration, or any task that benefits from parallel competition. Requires: a git repo.

Quality

78%

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 ./.gemini/skills/agenthub/SKILL.md

The canonical home for this skill is agenthub in alirezarezvani/claude-skills

Quality

Content

65%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 well-structured, concise, and action-oriented with concrete commands, but it inlines content that should be split into referenced files (some of which are referenced but missing) and lacks explicit pre-merge validation for a batch/destructive workflow.

Suggestions

Add an explicit validation checkpoint before /hub:merge — e.g., 'Verify the winning branch passes the full test suite and eval command before merging' — plus a conflict/rollback feedback loop, to satisfy the batch/destructive workflow requirement.

Move the agent-templates table, message-board post format, and DAG model details into separate files under references/ and scripts/, and ensure every referenced path (including references/agent-templates.md and the five scripts) actually exists as a bundle file.

Trim redundant restatements such as 'Agents do NOT see each other's work. They do NOT communicate...' which repeats the 'Agents work independently' claim from the intro.

Dimension	Reasoning	Score
Conciseness	Efficient use of tables, code blocks, and terse protocol descriptions without explaining concepts Claude already knows (no primer on git worktrees or DAGs); only minor padding like the redundant restating of agent isolation in the Agent Protocol section.	4 / 5
Actionability	Provides concrete, executable commands — 'python scripts/dag_analyzer.py --frontier --session {id}', 'python scripts/result_ranker.py --session {id} --eval-cmd "pytest bench.py --json" --metric p50_ms --direction lower', and 'git merge --no-ff' — with minor gaps in full flag signatures for some scripts.	4 / 5
Workflow Clarity	The lifecycle is clearly sequenced (INIT → DISPATCH → MONITOR → EVALUATE → MERGE with a state-machine table), but this is a batch and destructive operation (N agents, merge + archive + worktree cleanup) with no explicit validation checkpoint confirming the winner is correct before merging and no merge-conflict/rollback feedback loop, which caps the score per the batch-operation guideline.	3 / 5
Progressive Disclosure	Section structure with headers and tables is decent, but substantial content that belongs in separate files (agent templates, message-board post format, DAG model details) is inlined, and the one signaled reference 'references/agent-templates.md' points to a file that does not exist alongside absent scripts/, references/, and assets/ directories.	3 / 5
	Total	14 / 20 Passed

Description

92%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 capabilities, gives explicit 'Use when' trigger guidance, and carves out a distinctive niche. The only gap is slightly narrower trigger-term synonym coverage than the top anchor.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'spawns N parallel subagents competing on the same task via git worktree isolation', 'results are evaluated by metric or LLM judge', 'the best branch is merged' — giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both 'what' (spawns competing subagents, evaluates, merges winner) and 'when' via the concrete 'Use when: user wants multiple approaches tried in parallel ...' clause with specific trigger phrases.	5 / 5
Trigger Term Quality	Good natural trigger coverage with 'multiple approaches tried in parallel', 'code optimization', 'content variation', 'research exploration', and 'parallel competition', but misses common synonyms (fan-out, tournament, A/B test) that appear only in the body.	4 / 5
Distinctiveness Conflict Risk	Clear niche — parallel multi-agent competition with git worktree isolation and the 'Requires: a git repo' constraint — yielding distinct triggers and minimal overlap with other skills.	5 / 5
	Total	19 / 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
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: .gemini/skills/agenthub/SKILL.md
Commit: 19392f7

Reviewed: ~1 hour 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.