agent-agent

Agent skill for agent - invoke with $agent-agent

4.65x

Quality

13%

Does it follow best practices?

Impact

93%

4.65x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-agent/SKILL.md

Quality

Content

27%

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

The body is a sprawling, monolithic code dump that re-teaches familiar algorithms and relies on undefined helpers and inconsistently named MCP tools, so it is neither token-efficient nor reliably executable. It has a recognizable workflow spine but lacks genuine validation gates and offloads nothing to separate reference files.

Suggestions

Collapse the body to a lean overview and move the long code implementations into reference files under references/ (e.g. GOAP.md, examples.md), keeping only a concise quick-start and one-level-deep links.

Make code executable by defining or importing the helper functions and standardizing MCP tool names to match the listed tool inventory exactly.

Delete concept explanations Claude already knows (A*, PageRank, behavior trees) and the duplicate capability/usage sections, and add explicit validate-then-proceed checkpoints to the workflow.

Dimension	Reasoning	Score
Conciseness	The ~800-line body re-explains concepts Claude already knows (A* search, PageRank, behavior trees, utility selection) and is padded with redundant sections, buzzword capability lists, and five largely repetitive usage examples, matching the verbose/padded anchor at score 1 rather than the mostly-efficient score 2 anchor.	1 / 3
Actionability	Substantial code is present, but it is not executable: helper functions (canTransition, recursiveDecompose, buildDependencyMatrix, PriorityQueue, heuristic) are undefined, and MCP tool names are inconsistent (e.g. mcp__sublinear-time-solver__solve listed but mcp__sublinear_time_solver__solve called; mcp__flow-nexus__ vs mcp__claude_flow__), so it lands at score 2 for concrete-but-incomplete guidance rather than copy-paste-ready score 3.	2 / 3
Workflow Clarity	A numbered Workflow section (State Space Modeling through A* Search) provides a sequence and some validation touches (validateTemporalAdvantage, an error-handling section), but these are an architecture tour rather than explicit validation checkpoints with fix/retry feedback loops, so it sits at score 2 rather than the gated-checklist score 3.	2 / 3
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent) and the entire body is a monolithic inline wall of text with advanced config, examples, and code all stuffed into SKILL.md, matching the monolithic-anchor score 1 rather than the split-with-references score 2.	1 / 3
	Total	6 / 12 Passed

Description

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 frontmatter description is a non-functional placeholder that fails to convey what the skill does or when to invoke it, and is not distinguishable from other skills. The skill is also malformed: the meaningful description is buried in a second YAML block inside the body rather than the canonical frontmatter.

Suggestions

Replace the placeholder with a single frontmatter description that lists concrete GOAP actions (e.g., decomposes goals into action sequences, prioritizes via PageRank, replans on state changes) and remove the duplicate second YAML block.

Add an explicit "Use when..." trigger clause naming natural terms users would say, such as "complex multi-step objectives", "action planning", or "goal decomposition".

Use third-person active voice describing capabilities, and drop the "$agent-agent" invocation syntax that adds no user-facing signal.

Dimension	Reasoning	Score
Specificity	The description "Agent skill for agent - invoke with $agent-agent" names no concrete actions, only restating "agent" circularly and giving an invocation syntax, matching the vague/abstract anchor at score 1 rather than the score 2 anchor that names a domain and some actions.	1 / 3
Completeness	It answers neither "what does this do" (the phrasing is circular) nor "when should Claude use it" — there is no "Use when..." clause, and per the guidelines a missing explicit trigger caps completeness at 2, but here even the "what" is missing so it falls to score 1.	1 / 3
Trigger Term Quality	It contains no natural keywords a user would actually say; "agent skill for agent" and "$agent-agent" are generic/technical jargon with no common user-facing variations, so it cannot reach the score 2 anchor requiring some relevant keywords.	1 / 3
Distinctiveness Conflict Risk	"Agent skill for agent" is maximally generic and would conflict with virtually any other skill, matching the score 1 anchor rather than the score 2 anchor where some specificity narrows the overlap.	1 / 3
	Total	4 / 12 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
skill_md_line_count	SKILL.md is long (821 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: ruvnet/ruflo
Path: .agents/skills/agent-agent/SKILL.md
Commit: 26c35b5

Reviewed: about 6 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.