iterative-retrieval

Pattern for progressively refining context retrieval to solve the subagent context problem

Quality

35%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./docs/zh-TW/skills/iterative-retrieval/SKILL.md

Quality

Discovery

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This description is too abstract and jargon-heavy to be effective for skill selection. It reads more like an academic paper title than a functional skill description. It lacks concrete actions, natural trigger terms, and any explicit guidance on when Claude should select this skill.

Suggestions

Replace abstract language with concrete actions, e.g., 'Implements iterative context narrowing for subagent workflows by starting with broad searches and progressively filtering results to find relevant information.'

Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when a subagent needs to find specific context from a large codebase, when search results are too broad, or when dealing with multi-step retrieval tasks.'

Include natural keywords users might say, such as 'search refinement', 'finding relevant code', 'narrowing results', 'subagent search', or 'iterative lookup'.

Dimension	Reasoning	Score
Specificity	The description uses abstract language like 'pattern for progressively refining context retrieval' without listing any concrete actions. It describes a concept rather than specific capabilities.	1 / 3
Completeness	The description weakly addresses 'what' (a pattern for refining context retrieval) and completely lacks any 'when' clause or explicit trigger guidance for when Claude should use this skill.	1 / 3
Trigger Term Quality	The terms 'progressively refining context retrieval' and 'subagent context problem' are technical jargon that users would not naturally say. There are no natural trigger keywords a user would use.	1 / 3
Distinctiveness Conflict Risk	The mention of 'subagent context problem' is somewhat niche and specific to a particular domain, which reduces conflict risk slightly, but the overall vagueness of 'context retrieval' could overlap with many retrieval-related skills.	2 / 3
	Total	5 / 12 Passed

Implementation

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill clearly articulates a useful pattern for iterative context retrieval with strong workflow clarity and good practical examples. Its main weaknesses are that the code examples are conceptual rather than executable, and the content is somewhat verbose for what could be a more concise pattern description. The problem statement section over-explains things Claude already understands.

Suggestions

Make code examples more actionable by either providing truly executable implementations or reframing them as concrete agent prompt instructions rather than abstract function signatures with undefined helpers.

Trim the 'Problem' section significantly—Claude already understands subagent context limitations; a single sentence would suffice.

Consider splitting the two detailed practical examples into a separate EXAMPLES.md file to keep the main skill lean and improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The skill includes some unnecessary verbosity—the ASCII diagram is nice but the problem description section explains things Claude already understands (what subagents lack, why standard approaches fail). The code examples are somewhat illustrative rather than minimal, and the dual practical examples add length. However, it's not egregiously padded.	2 / 3
Actionability	The code examples are illustrative pseudocode/conceptual JavaScript rather than truly executable, copy-paste-ready code. Functions like `scoreRelevance`, `explainRelevance`, `retrieveFiles` are undefined abstractions. The practical examples (bug fix, feature implementation) are helpful walkthroughs but remain conceptual patterns rather than concrete executable guidance.	2 / 3
Workflow Clarity	The 4-stage cycle (DISPATCH → EVALUATE → REFINE → LOOP) is clearly sequenced with explicit stopping conditions (3 cycles max, relevance >= 0.7, hasCriticalGaps check). The two practical examples demonstrate the workflow concretely with clear iteration and termination criteria. The feedback loop (evaluate → refine → retry) is well-defined.	3 / 3
Progressive Disclosure	The content is structured with clear sections and headers, but it's a monolithic document with ~150 lines of inline content that could benefit from splitting (e.g., examples and integration guidance into separate files). The 'Related' section references external resources but the links are somewhat vague (an X/Twitter link, a skill name without clear path). No bundle files support the references.	2 / 3
	Total	9 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: affaan-m/everything-claude-code
Commit: 928076c

Reviewed: 3 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.