flow-deliver

Multi-AI validation, scoring, and review using available external providers (Double Diamond Deliver phase)

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/flow-deliver/SKILL.md

Quality

Content

62%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 strong on workflow clarity (enforced gated sequence with validation and feedback loops) and reasonably actionable, but it is severely over-long due to tripled workflow descriptions and a huge fabricated example report, with no progressive disclosure via reference files. The dominant weakness is conciseness from heavy duplication.

Suggestions

Collapse the three repeated presentations of the workflow (EXECUTION CONTRACT, "Deliver Workflow - Deliver Phase", and "How It Works") into a single canonical sequenced flow; keep one banner template instead of three.

Move the ~130-line Example 1 report and the validation report format template into a reference file (e.g. references/report-format.md) and link to it, improving both conciseness and progressive disclosure.

Extract the dev-subtype validation supplement table into references/validation-supplements.md so the SKILL.md body stays a lean overview.

Dimension	Reasoning	Score
Conciseness	The ~885-line body noticeably duplicates the workflow across the EXECUTION CONTRACT, a second "Deliver Workflow" section, and a "How It Works" section, repeats the provider banner three times, and inlines a ~130-line fabricated example report — several padded sections that should be trimmed, matching the anchor for noticeably verbose content.	2 / 5
Actionability	Concrete, copy-paste-ready bash commands (orchestrate.sh deliver, check-providers.sh, state-manager.sh, validation-file checks, gh PR posting) and a subtype table with specific validation criteria provide mostly executable guidance, with only minor templated placeholders like <user's validation request>.	4 / 5
Workflow Clarity	The enforced 7-step contract is explicitly sequenced with blocking "DO NOT PROCEED" gates, an explicit validation gate in Step 5 (verify the validation file exists or exit 1), feedback loops in the Error Handling section, and a terminal validation checklist — matching the anchor for clear sequence with explicit validation, feedback loops, and checklists.	5 / 5
Progressive Disclosure	Section headers provide some structure, but no bundle reference files exist (references/, scripts/, assets/ absent) and content that clearly belongs in separate files — the giant example report, the validation report format template, and the repeated banners — is all inlined, matching the anchor for some structure with content that should be separate kept inline.	3 / 5
	Total	14 / 20 Passed

Description

61%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 concise and names a distinct multi-AI review niche with natural trigger terms, but it omits any explicit "Use when..." guidance in the description field itself, capping completeness. Specificity is moderate because the named actions are abstract synonyms rather than concrete operations.

Suggestions

Add an explicit "Use when..." clause to the description (e.g., "Use when the user asks to review, validate, score, or quality-check code or documents before shipping") so completeness can reach 4-5 without relying on the separate trigger field.

Replace the abstract verb list "validation, scoring, and review" with more concrete capabilities (e.g., "runs multi-provider code review, security audit, and quality scoring") to lift specificity.

Include common trigger synonyms (test, check, audit, quality) in the description to push trigger-term coverage toward 5.

Dimension	Reasoning	Score
Specificity	Names the domain ("Multi-AI") and three actions ("validation, scoring, and review"), but these are abstract near-synonyms rather than distinct concrete operations, matching the anchor that lists 1-2 concrete actions without comprehensive coverage.	3 / 5
Completeness	The description gives a clear "what" (Multi-AI validation, scoring, review) but no "when"/"Use when" clause — the trigger guidance lives in a separate `trigger:` field, so per the rubric a missing explicit trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	"validation, scoring, and review" are three natural terms a user would say ("review this", "score it", "validate before shipping"), giving good keyword coverage with only a few synonyms missing (test, check, audit, quality), matching the anchor for good coverage with minor gaps.	4 / 5
Distinctiveness Conflict Risk	The "Multi-AI" + "available external providers" + "Double Diamond Deliver phase" framing carves a mostly distinct niche, though the broad "validation/scoring/review" verbs carry minor overlap risk with general review skills.	4 / 5
	Total	14 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (926 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-deliver/SKILL.md
Commit: 805d58c

Reviewed: 24 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.