iterative-retrieval

サブエージェントのコンテキスト問題を解決するために、コンテキスト取得を段階的に洗練するパターン

Quality

54%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./docs/ja-JP/skills/iterative-retrieval/SKILL.md

Quality

Content

67%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 content is a well-structured, actionable walkthrough of the iterative-retrieval pattern with executable code, a clear phased workflow, and explicit validation checkpoints. Its main weaknesses are mild verbosity in the framing/diagram and reliance on several undefined helper functions that keep the code from being fully copy-paste ready.

Suggestions

Trim the 'problem' section and the ASCII diagram to the essentials, or move the diagram behind a short bullet summary, to reduce token overhead.

Provide minimal definitions (or signatures with expected return shapes) for the helper functions like retrieveFiles, scoreRelevance, extractPatterns, and mergeContext so the code is runnable as written.

Consider splitting the two long worked examples into a separate EXAMPLES.md reference so the main body stays a concise overview with one-level-deep navigation.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with lean code blocks, but the 'problem' framing, the decorative ASCII diagram, and the worked examples that re-walk the same four phases add tokens that could be tightened without losing clarity.	3 / 5
Actionability	It provides concrete JavaScript for all four phases plus two fully worked examples, but the code calls many undefined helpers (retrieveFiles, scoreRelevance, extractPatterns, mergeContext, hasCriticalGaps), so it is mostly executable with minor gaps rather than copy-paste ready.	4 / 5
Workflow Clarity	The DISPATCH->EVALUATE->REFINE->LOOP sequence is clearly laid out with an explicit stop condition (relevance >= 0.7, >= 3 files, no critical gaps) and a refine-and-retry feedback loop, missing only a true checklist to reach 5.	4 / 5
Progressive Disclosure	The skill is a single self-contained, well-sectioned document (no nested file references; the related-items links are one level deep), with good organization and only minor room to split the worked examples out.	4 / 5
	Total	15 / 20 Passed

Description

41%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 but omits any explicit 'when to use it' trigger guidance and relies on technical jargon rather than natural user phrases. It is reasonably distinct but would benefit from a concrete 'Use when...' clause with everyday trigger terms.

Suggestions

Add an explicit 'Use when...' clause naming natural trigger phrases, e.g. 'Use when a sub-agent needs relevant files/context but cannot determine them upfront, or when orchestrating multi-agent retrieval.'

Replace abstract phrasing ('progressively refine context retrieval') with a couple of concrete verbs describing what the agent actually does (e.g. 'iteratively search, score, and narrow candidate files').

Include common user-facing synonyms such as 'context gathering', 'relevant files', 'retrieval loop' alongside the technical terms to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ('sub-agent context problem') and a single fairly abstract action ('progressively refine context retrieval'), but frames it as a 'pattern' rather than listing concrete actions, so it sits at the minimal/generic-actions level rather than the 1-2 concrete actions level.	2 / 5
Completeness	It gives a clear 'what' (a pattern for progressively refining context retrieval) but provides no explicit 'Use when...' trigger clause, so the 'when' is only weakly implied via 'to solve the sub-agent context problem', capping it at 3 per the missing-trigger guideline.	3 / 5
Trigger Term Quality	It contains a couple of recognizable keywords ('sub-agent', 'context') but the phrasing ('progressively refine context retrieval', 'context problem') is technical jargon rather than the natural phrases a user would say when they need this skill.	2 / 5
Distinctiveness Conflict Risk	The niche (sub-agent context retrieval refinement) is fairly specific and distinguishable from most skills, with only minor overlap risk against other agent-orchestration skills, matching the 'mostly distinct' anchor.	4 / 5
	Total	11 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: affaan-m/ECC
Path: docs/ja-JP/skills/iterative-retrieval/SKILL.md
Commit: e04ea0b

Reviewed: 2 days 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.