plannotator-compound

Analyze a user's Plannotator plan archive to extract denial patterns, feedback taxonomy, evolution over time, and actionable prompt improvements — then produce a polished HTML dashboard report. Falls back to Claude Code ExitPlanMode denial reasons when Plannotator data is unavailable.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

85%

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

A highly actionable, well-sequenced multi-phase workflow that uses its bundle files effectively for progressive disclosure. The main weakness is conciseness: the inlined dashboard design spec and prompt templates make the body longer than necessary.

Suggestions

Move the Phase 4 dashboard design language and section-by-section HTML spec into a reference file (e.g. references/dashboard-design.md) and keep only the section order and key rules inline.

Extract the full extraction-prompt and reduction-prompt templates into a reference file, leaving short summaries and the variable adaptations inline so the core flow stays scannable.

Tighten repeated fallback-mode notes by consolidating them into the existing claude-code-fallback.md reference and linking once per phase instead of restating.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient operational guidance with no general-knowledge padding, but the ~130-line Phase 4 dashboard design spec and the fully inlined extraction/reduction prompts could be tightened or offloaded, so it does not reach the lean score-3 anchor.	2 / 3
Actionability	Provides concrete shell commands, exact output paths, full copy-paste agent prompt templates, dataset-size batching thresholds, and model-selection directives — fully executable guidance rather than vague direction.	3 / 3
Workflow Clarity	Phases 0–6 are explicitly sequenced with a hard barrier ('Each phase must complete fully before the next begins') and validation checkpoints — count verification against the Phase 1 inventory and a timeout re-launch feedback loop for batch agents.	3 / 3
Progressive Disclosure	Three real, one-level-deep bundle files (references/claude-code-fallback.md, scripts/extract_exit_plan_mode_outcomes.py, assets/report-template.html) are clearly signaled with markdown links and split out appropriately, leaving the body as a navigable overview.	3 / 3
	Total	11 / 12 Passed

Description

82%

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

A specific, well-scoped description with concrete actions and clear distinctiveness, but it is missing an explicit 'Use when…' usage trigger, which caps completeness. Adding a trigger clause would round it out fully.

Suggestions

Add an explicit 'Use when…' clause, e.g. 'Use when the user wants to analyze their denied Plannotator plans or Claude Code ExitPlanMode denials and turn recurring feedback into prompt improvements.'

Surface the most common trigger phrasings users would actually type ('denied plans', 'plan review feedback', 'why my plans get rejected') so the description matches natural requests.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'extract denial patterns, feedback taxonomy, evolution over time, and actionable prompt improvements' and 'produce a polished HTML dashboard report' — matching the score-3 anchor of several specific actions.	3 / 3
Completeness	It clearly answers 'what' but lacks an explicit 'Use when…' trigger clause; the 'Falls back… when Plannotator data is unavailable' sentence describes fallback behavior, not a usage trigger, so per the guideline completeness is capped at 2.	2 / 3
Trigger Term Quality	Natural terms a target user would say — 'Plannotator plan archive', 'denial patterns', 'feedback taxonomy', 'HTML dashboard report' — give good coverage, and 'ExitPlanMode' is the legitimate domain term for the fallback.	3 / 3
Distinctiveness Conflict Risk	The Plannotator / Claude Code plan-denial analysis niche with its specific triggers is clearly distinct and unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 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
skill_md_line_count	SKILL.md is long (576 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: backnotprop/plannotator
Path: apps/skills/extra/plannotator-compound/SKILL.md
Commit: 193b07e

Reviewed: 6 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.