CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-v3-integration-architect

Agent skill for v3-integration-architect - invoke with $agent-v3-integration-architect

51

4.42x
Quality

26%

Does it follow best practices?

Impact

93%

4.42x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-v3-integration-architect/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

25%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.

This reads as a project status document rather than an actionable skill: heavily duplicated targets, speculative unverifiable numbers (150x-12,500x, <0.05ms), and non-executable pseudocode throughout. An agent following it would not know what command to run or how to verify a migration phase succeeded before deleting code. It needs de-duplication, executable commands, and validation gates.

Suggestions

Consolidate the triplicated performance targets and overlap percentages into one table; cut the stray second YAML frontmatter block at the top of the body entirely.

Replace the invented TypeScript class pseudocode with real, runnable commands (e.g. actual npx agentic-flow@alpha invocations and file-removal steps) so each phase is executable.

Add explicit validation checkpoints to the destructive Phase 3: run a parity/benchmark check after each migration and only delete the deprecated file (SwarmCoordinator.ts, AgentManager.ts, TaskScheduler.ts) after the check passes.

Move the detailed migration plan, benchmark tables, and agent coordination notes into references/ files (e.g. references/migration-plan.md, references/benchmarks.md), leaving SKILL.md as a lean overview with clearly signaled one-level-deep links.

DimensionReasoningScore

Conciseness

The body duplicates the same information repeatedly: the overlap percentages appear in the pre_execution hook echoes AND the ASCII table AND the success-metrics checklist; the Flash Attention "2.49x-7.47x" and AgentDB "150x-12,500x" targets each appear three times. ~300 lines where a quarter would do, plus a stray duplicated YAML frontmatter block at the top of the body. Not score 1 (there is no padding explaining concepts Claude already knows), but it is noticeably verbose with several redundant sections.

2 / 5

Actionability

All code is TypeScript pseudocode against invented APIs (this.agenticFlow.sona.setMode, this.removeFile(...), this.agentdb.enableCrossAgentSharing) that cannot execute; even the hook shell commands are garbled ("2>$dev$null"). There is high-level guidance and named migration steps, but nothing copy-paste ready — matching anchor 2 (high-level hints missing specific executable steps), and clearly not anchor 3's partial-executability or 4's runnable code.

2 / 5

Workflow Clarity

A rough phase sequence exists (Phase 1 adapter, Phase 2 migration, Phase 3 removal), but no validation checkpoints anywhere: Phase 3 deletes files (src/core/SwarmCoordinator.ts, src/agents/AgentManager.ts) with no verify-parity-before-removal step, and the 'validateFeatureParity' call is invented pseudocode with no command behind it. Guideline: a destructive/batch workflow without validation caps at 3, and this is below that cap anyway — steps are poorly defined with many gaps (anchor 2), not the listed-but-checkpoint-missing state of anchor 3.

2 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent) and the body references none. Everything — the duplication analysis, the three-phase migration plan, benchmark tables, and coordination notes — is inlined monolithically when the migration plan and benchmarks clearly belong in separate reference files. Anchor 2 ("content that clearly belongs in separate files is inlined") fits; not 1 because section headers do provide navigable structure, and not 3 because there are no references at all to be 'present but unclear'.

2 / 5

Total

8

/

20

Passed

Description

28%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 effectively a placeholder: it restates the skill name and an invocation hint without saying what the skill does or when to use it. It would rarely be surfaced by a real user request and fails both the 'what' and 'when' requirements. This is a weak description requiring a full rewrite.

Suggestions

Replace the description with concrete third-person actions, e.g. "Implements ADR-001 by migrating claude-flow components (SwarmCoordinator, AgentManager, TaskScheduler, SessionManager) to agentic-flow@alpha, eliminating 10,000+ duplicate lines while preserving feature parity."

Add an explicit trigger clause: "Use when migrating claude-flow modules to agentic-flow, deduplicating orchestration code, or working on ADR-001 integration phases."

Drop the "invoke with $agent-v3-integration-architect" meta-instruction — it is invocation plumbing, not capability description, and adds no trigger value.

DimensionReasoningScore

Specificity

The description only states "Agent skill for v3-integration-architect", which names a domain-ish title but describes zero concrete actions — no verbs, no capabilities. It is above anchor 1 (it names a specific role/domain rather than pure abstraction like "Helps with documents") but well below anchor 3, which requires 1-2 concrete actions.

2 / 5

Completeness

The 'what' is vague ("Agent skill for..." — what does it actually do?) and the 'when' is entirely absent; there is no 'Use when...' clause or equivalent, which caps this dimension at 3 per the guidelines. This matches anchor 2 ("vague 'what' and no 'when'") exactly, not anchor 3 which requires a clear 'what'.

2 / 5

Trigger Term Quality

The only keywords are the skill's own name ("v3-integration-architect", "invoke with $agent-v3-integration-architect") plus "integration"/"architect" — jargon a user would not naturally say when needing this skill. Not score 1 because it does contain one or two domain terms; not 3+ because no natural trigger phrases or synonyms appear.

2 / 5

Distinctiveness Conflict Risk

The narrowly-scoped name (v3 integration, ADR-001, agentic-flow) makes it unlikely to collide with generic skills, but the description gives no distinguishing trigger information, so it could overlap with other migration/architecture skills in the same project. Between anchor 2 (very broad, high overlap) and anchor 4 (mostly distinct with clear triggers); it is somewhat specific without being clearly signaled.

3 / 5

Total

9

/

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.