humanize-refine-plan

Refine an annotated implementation plan into a comment-free plan and a QA ledger while preserving the gen-plan schema.

Quality

65%

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 ./skills/humanize-refine-plan/SKILL.md

Quality

Content

62%Weight 40%Scale 1-3

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

The body presents a clear, well-validated workflow with concrete CLI hooks and a strong feedback loop, but it is somewhat redundant between the flowchart and prose, describes the core refinement logic abstractly, and keeps all detail inline with no progressive disclosure to reference files. Tightening the chart, making the processing steps more executable, and splitting the lookup tables would raise the lower dimensions.

Suggestions

Reduce redundancy between the mermaid flowchart node labels and the Workflow Guarantees / Classification prose so the chart shows sequence while the prose adds only what the chart cannot.

Make the core refinement steps more executable by specifying how CMT/ENDCMT blocks are scanned and classified (e.g. a concrete scanner rule or minimal code) instead of describing them abstractly.

Move the exit-code and alternate-language lookup tables into reference files (or clearly signal them as one-level-deep references) to introduce genuine progressive disclosure and slim the main SKILL.md.

Dimension	Reasoning	Score
Conciseness	The body is lean and avoids teaching known concepts, but the mermaid flowchart node labels restate much of the subsequent Workflow Guarantees and Classification prose, so it is mostly efficient yet could be tightened rather than earning the every-token-counts anchor at 3.	2 / 3
Actionability	Concrete CLI invocations (validate-refine-plan-io.sh call, /flow and /skill usage), an exit-code table, and flag defaults are present, but the core refinement logic is descriptive ("Process comments in order: answer, refine plan, or do targeted repo research") rather than executable, landing at some-concrete-guidance-but-incomplete rather than copy-paste-ready at 3.	2 / 3
Workflow Clarity	The flowchart lays out a clear sequence with explicit validation (VALIDATE, VALID_OK), a plan-check feedback loop (PLAN_CHECK to FIX and re-check), blocking-vs-fixable branches, atomic writes, and documented exit codes, matching the explicit-checkpoints-with-feedback-loops anchor rather than the implicit-checkpoint anchor at 2.	3 / 3
Progressive Disclosure	Sections are well organized, but no bundle files exist and the only external reference is a runtime-templated QA template path; content that could be split (exit-code and language-variant tables) is inline, so it sits at structured-but-inline rather than the well-signaled one-level-deep reference anchor at 3.	2 / 3
	Total	9 / 12 Passed

Description

67%Weight 40%Scale 1-3

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 specific and distinctive, naming concrete outputs and a clear niche, but it omits any explicit "Use when..." trigger guidance and leans on technical jargon over natural user phrasing. Adding a trigger clause and a few plain-language terms would lift completeness and trigger-term quality.

Suggestions

Append an explicit trigger clause such as "Use when refining an annotated gen-plan plan or stripping CMT comment blocks from a plan" to satisfy the "when" half of completeness.

Soften jargon by adding natural terms users would actually say (e.g. "remove comments from a plan", "clean up an annotated plan") alongside "gen-plan schema" and "QA ledger".

Keep the concrete action list but pair it with the trigger phrasing so distinctiveness is reinforced by user-facing language rather than only internal schema terms.

Dimension	Reasoning	Score
Specificity	"Refine an annotated implementation plan into a comment-free plan and a QA ledger while preserving the gen-plan schema" lists several concrete actions and outputs (refine, produce comment-free plan, produce QA ledger, preserve schema), matching the multiple-specific-actions anchor rather than the partial-domain anchor at 2.	3 / 3
Completeness	It clearly answers "what" the skill does but provides no "Use when..." clause or equivalent explicit trigger for "when" to use it, which per the guidelines caps completeness at 2 rather than reaching the explicit-trigger anchor at 3.	2 / 3
Trigger Term Quality	Terms like "annotated implementation plan", "QA ledger", and "gen-plan schema" are specialized jargon and miss the natural phrasings a user would actually say (e.g. "clean up comments in my plan"), so it sits at "some relevant keywords but missing common variations" rather than full natural-term coverage at 3.	2 / 3
Distinctiveness Conflict Risk	The scope is tightly bound to a specific niche (annotated gen-plan plans converted to a comment-free plan plus QA ledger), making it unlikely to trigger for unrelated skills, matching the clear-niche anchor rather than the overlapping-domain anchor at 2.	3 / 3
	Total	10 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: PolyArch/humanize
Path: skills/humanize-refine-plan/SKILL.md
Commit: 0ec921a

Reviewed: 8 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.