CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

flow-develop

Multi-AI implementation using available external providers (Double Diamond Develop phase). DO NOT use for simple code edits, reading/reviewing code, built-in commands, or trivial single-file changes.

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/flow-develop/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

48%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 is highly actionable with concrete commands, explicit validation gates, and strong error-handling guidance, but it is crippled by severe redundancy — the same context-detection and banner instructions appear three times with conflicting step numbering — and by its monolithic single-file structure. Consolidating the duplicated workflow descriptions would cut the file roughly in half while improving clarity.

Suggestions

Eliminate the duplicated context-detection, subtype, and banner sections: keep the EXECUTION CONTRACT as the single authoritative sequence and remove the repeated 'MANDATORY: Context Detection & Visual Indicators', 'Visual Indicators', and 'How It Works' restatements.

Move the subtype quality-supplement table, frontend-ui enrichment details, agent-continuation protocol, and worked examples into reference files (e.g., references/subtypes.md, examples.md) linked one level deep from a concise overview.

Isolate version-sensitive notes (v7.16.0+ spinner behavior, CC v2.1.77+ SendMessage changes) into a clearly labeled compatibility section so they don't penalize the main instructions.

Replace name-only invocations of bridge_store_agent_id and skill-verification-gate with their concrete call syntax or a reference to where it is defined.

DimensionReasoningScore

Conciseness

The ~830-line body is severely padded: context detection, subtype logic, and provider banners are each repeated two to three times (EXECUTION CONTRACT, 'MANDATORY: Context Detection & Visual Indicators', 'Visual Indicators'), with ASCII art, repeated PROHIBITED/WHY exhortations, and version-sensitive details (v7.16.0+, CC v2.1.16+, v9.5 / CC v2.1.77+) not isolated in a deprecated section. This matches anchor 1, 'severely verbose... heavily padded'.

1 / 5

Actionability

The body provides copy-paste-ready bash for orchestrate.sh, state-manager.sh, check-providers.sh, and feature-workflow.sh, plus full agent-launch specs and a worked auth-system example. Minor gaps keep it below anchor 5: bridge_store_agent_id and skill-verification-gate are invoked by name without syntax, and the BM25 snippet contains unfilled template placeholders.

4 / 5

Workflow Clarity

The EXECUTION CONTRACT's Steps 1-7 are clearly sequenced with mandatory blocking gates, explicit validation (Step 5 report verification), error-handling feedback loops, and before/after checklists. It falls short of anchor 5 because a parallel 'How It Works' section restates the same workflow as Steps 1-5 with different numbering and ordering, creating conflicting sequences.

4 / 5

Progressive Disclosure

No bundle files exist, so everything — the subtype table, agent-continuation protocol, banner templates, and worked examples — is inlined in one ~830-line file. Section headers do provide navigable structure (ruling out anchor 2's 'minimal structure'), but substantial content that belongs in separate reference files is inline, fitting anchor 3.

3 / 5

Total

12

/

20

Passed

Description

57%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 draws sharp negative boundaries, which supports distinctiveness, but it lacks concrete capability details and any positive trigger phrases. Users saying "build X" or "develop a feature" would not be captured by the current wording.

Suggestions

Add a positive trigger clause such as 'Use when the user asks to build, implement, or develop a feature or system' to complement the DO-NOT list.

List 2-3 concrete actions, e.g., 'generates multiple implementation approaches via external providers, synthesizes them, and runs quality gates before implementing'.

Include natural synonyms (build, implement, develop, create) in the trigger language so common user phrasings match.

DimensionReasoningScore

Specificity

"Multi-AI implementation using available external providers" names the domain but offers only the generic action "implementation" — no concrete actions (e.g., generating multiple approaches, synthesizing provider outputs, quality gates) are stated. It matches the anchor 'names the domain but actions are minimal or generic' rather than anchor 3, which requires 1-2 concrete named actions.

2 / 5

Completeness

The 'what' is clearly stated (multi-AI implementation via external providers) and the 'when' is explicitly present in negative form ("DO NOT use for simple code edits, reading/reviewing code, built-in commands, or trivial single-file changes"). This fits anchor 4 — both present with the 'when' needing more explicit positive triggers — rather than anchor 3, where 'when' is missing or merely implied.

4 / 5

Trigger Term Quality

"implementation" plus natural exclusion phrases like "simple code edits" and "reading/reviewing code" give some relevant keywords, but common positive triggers users actually say ("build", "develop", "create", "feature") are absent. Coverage is partial, fitting anchor 3 rather than anchor 4's 'good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

The niche (multi-AI implementation, Double Diamond Develop phase) is distinct and the DO-NOT list explicitly carves it away from plain editing, reading, and built-in commands, giving minor overlap risk with closely related build/code skills. Not anchor 5 because "implementation" alone still broadly overlaps general coding skills without positive scoping.

4 / 5

Total

13

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (871 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

nyldn/claude-octopus

Path

.claude/skills/flow-develop/SKILL.md

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.