CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-consensus-coordinator

Agent skill for consensus-coordinator - invoke with $agent-consensus-coordinator

52

7.38x
Quality

26%

Does it follow best practices?

Impact

96%

7.38x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-consensus-coordinator/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

32%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 presents a broad but shallow catalog: real MCP tool names are its main asset, yet all code is non-executable pseudocode with undefined helper methods, workflows lack any validation steps, and half the document restates distributed-consensus concepts Claude already knows. It also opens with a stray duplicate frontmatter block, a formatting defect.

Suggestions

Cut the concept-catalog sections ('Advanced Consensus Algorithms', 'Performance Optimization', 'Fault Tolerance Mechanisms') that merely list pBFT/PoS/CAP-theory terminology, and keep only skill-specific guidance such as the MCP tool parameter reference.

Make the code examples executable by defining (or removing) the undefined helpers like buildConsensusMatrix/extractAgreement, and fix the broken require('.$consensus-network') path.

Add explicit validation checkpoints to the workflows (e.g. verify analyzeMatrix's spectralGap before declaring Byzantine resilience, and re-check after topology changes), and move the lengthy Flow Nexus deployment examples into a reference file once one exists.

DimensionReasoningScore

Conciseness

Roughly half the ~340-line body is padded bullet-list sections that catalog concepts Claude already knows — 'Practical Byzantine Fault Tolerance (pBFT): Three-Phase Protocol', 'Proof of Stake Consensus: Slashing Conditions', 'CAP Theorem Optimization', 'Sharding' — with no skill-specific detail added. It is below anchor 3 ('some unnecessary explanation') but stops short of anchor 1's pure tutorial prose because the MCP tool inventory and code examples do carry information.

2 / 5

Actionability

There is concrete guidance — real MCP tool names with parameters ('mcp__sublinear-time-solver__solve' with method/epsilon/maxIterations, 'analyzeMatrix', 'pageRank') — but every code example is pseudocode relying on undefined helpers ('this.buildConsensusMatrix', 'this.extractAgreement', 'this.calculateReliability') and includes a broken require ('.$consensus-network'). This matches anchor 3 ('pseudocode instead of executable code; missing key details'), not anchor 4 which requires executable code with only minor gaps.

3 / 5

Workflow Clarity

The 'Example Workflows' sections give rough 5-step sequences ('Network Design' → 'Protocol Selection' → ... → 'Monitoring') but the steps are generic labels with no commands, and there are no validation checkpoints anywhere for operations that deploy consensus infrastructure and detect Byzantine nodes. This matches anchor 2 ('rough sequence present but many gaps; validation absent') and cannot reach anchor 3-4 without at least implicit verification steps.

2 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are all absent) and the body references no external files at all — everything, including full API-style code examples, is inlined in one long document. Section headers exist (better than a monolithic wall), which keeps it above anchor 1, but content that clearly belongs in separate reference files (the Flow Nexus deployment code, the algorithm catalogs) is fully inlined, matching anchor 2.

2 / 5

Total

9

/

20

Passed

Description

21%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 boilerplate wrapper text that fails to communicate what the skill does or when to use it. The actual capability vocabulary (distributed consensus, Byzantine fault tolerance, voting, sublinear solvers) exists only in the body's stray second frontmatter block, which the description field never surfaces.

Suggestions

Replace the wrapper with a capability statement in third person, e.g. 'Designs and implements distributed consensus protocols (Byzantine fault tolerance, voting mechanisms, multi-agent coordination) using sublinear solver MCP tools.'

Add an explicit 'Use when...' trigger clause naming natural user phrases such as 'consensus protocol', 'Byzantine agreement', 'distributed voting', or 'multi-agent coordination'.

Remove the duplicated/stray second frontmatter block in the body and consolidate its richer description into the single real frontmatter description field.

DimensionReasoningScore

Specificity

The description 'Agent skill for consensus-coordinator - invoke with $agent-consensus-coordinator' names a domain but contains zero action verbs or concrete capabilities. It sits below anchor 2's example ('Processes PDF files' at least names an action) but above anchor 1 (it is not pure abstraction — a domain is named), so it falls between the anchors with no capability content at all.

2 / 5

Completeness

The 'what' is vague ('Agent skill for consensus-coordinator' states a category, not a capability) and the 'when' is entirely absent — 'invoke with $agent-consensus-coordinator' is an invocation instruction, not a 'Use when...' clause. This matches anchor 2 ('vague what and no when'); it is not anchor 3 because no clear, action-based 'what' is ever stated, and the missing-when rule would cap completeness at 3 regardless.

2 / 5

Trigger Term Quality

The only terms present are the technical identifiers 'consensus-coordinator' and '$agent-consensus-coordinator'; there are no natural phrases a user would say such as 'distributed consensus', 'voting', or 'Byzantine fault tolerance'. This matches anchor 1 ('only technical jargon or entirely generic language') — the richer trigger vocabulary exists only in the body, not the description.

1 / 5

Distinctiveness Conflict Risk

'consensus-coordinator' points at a genuinely niche domain, so it is not broadly generic (not anchor 1-2), but the boilerplate 'Agent skill for X - invoke with $X' wrapper carries no differentiating detail and would look identical to any sibling skill generated the same way, so overlap risk remains — anchor 3 ('somewhat specific but could still overlap').

3 / 5

Total

8

/

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
ruvnet/ruflo
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.