Skill de gerenciamento de contexto e tarefas usando o mecanismo de task, memoria ou checklist disponivel no ambiente, com persistencia enxuta entre sessoes quando fizer sentido. Use quando precisar organizar tarefas, acompanhar progresso, trocar foco de trabalho, ou manter historico resumido. Trigger em: "criar tarefa", "lista de tarefas", "status", "progresso", "trocar foco", "resetar contexto", "o que falta", "proxima tarefa", "historico", "pipeline".
67
58%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/08-context-manager/SKILL.mdQuality
Discovery
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.
This is a reasonably well-crafted description that clearly communicates both purpose and trigger conditions, with an explicit list of natural trigger terms in Portuguese. Its main weaknesses are moderate specificity in describing concrete actions (it stays at a somewhat abstract level) and some risk of overlap with other task/project management skills. The explicit trigger term list is a strong point that aids skill selection.
Suggestions
Add more concrete, specific actions to improve specificity, e.g., 'cria e atualiza checklists em arquivos markdown, persiste estado em .task files, gera resumos de progresso'.
Differentiate more clearly from generic project management or issue-tracking skills by specifying the unique mechanism used (e.g., naming the specific file format or tool integration for persistence).
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (context and task management) and mentions some actions like organizing tasks, tracking progress, switching focus, and maintaining summarized history. However, the actions are somewhat general and not as concrete as listing specific operations (e.g., 'create checklists', 'persist state to files', 'generate progress reports'). | 2 / 3 |
Completeness | The description clearly answers both 'what' (context and task management with persistence between sessions) and 'when' (explicitly states 'Use quando precisar organizar tarefas, acompanhar progresso, trocar foco de trabalho, ou manter historico resumido' plus a dedicated trigger list). Both dimensions are explicitly addressed. | 3 / 3 |
Trigger Term Quality | The description explicitly lists natural trigger terms in Portuguese that users would actually say: 'criar tarefa', 'lista de tarefas', 'status', 'progresso', 'trocar foco', 'resetar contexto', 'o que falta', 'proxima tarefa', 'historico', 'pipeline'. These are concrete, varied, and cover common user phrasings well. | 3 / 3 |
Distinctiveness Conflict Risk | While the description carves out a niche around task/context management with session persistence, terms like 'tarefa', 'status', and 'progresso' could overlap with project management, issue tracking, or general productivity skills. The mention of 'checklist' and 'memoria' mechanisms adds some specificity but the overall scope is broad enough to risk conflicts with similar organizational skills. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads more like a role description or process document than an actionable skill for Claude. It lacks concrete examples, executable commands, or templates that would make it immediately usable. The structure is reasonable but the content is too abstract—it describes what to do conceptually without showing how to do it with specific tool calls or output formats.
Suggestions
Add concrete, executable examples showing actual tool calls or file content for creating tasks, updating status, and persisting context (e.g., a sample current-focus.md template or a specific TodoWrite invocation).
Include a copy-paste-ready template for the handoff summary and the history.md file format so Claude knows exactly what output to produce.
Add a validation/verification step in the task lifecycle (e.g., 'after marking completed, verify the task list is consistent and no dependencies are broken') to strengthen the workflow.
Consolidate overlapping sections (Entradas/Saidas/Responsabilidades/Evidencia) into a single concise workflow with inline expectations rather than separate abstract lists.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably organized but includes some sections that are somewhat redundant or could be tightened (e.g., 'Quando Usar' vs 'Quando Nao Usar' overlap with 'Responsabilidades'; 'Entradas Esperadas' and 'Saidas Esperadas' are somewhat generic). The governance preamble and policy references add overhead without concrete value in the body itself. | 2 / 3 |
Actionability | The content is almost entirely descriptive and procedural in the abstract—there are no concrete code examples, specific commands, tool invocations, or copy-paste-ready templates. The task lifecycle is described as a text arrow diagram rather than executable steps with actual tool usage. Phrases like 'usar o mecanismo disponivel no ambiente' are vague. | 1 / 3 |
Workflow Clarity | There is a recognizable lifecycle (pending → in_progress → completed) and a focus-change workflow with steps, but validation checkpoints are missing. There's no explicit verification step after task updates, no error recovery loop, and the 'Ciclo de Vida' is a single-line arrow rather than a sequenced checklist with checks. | 2 / 3 |
Progressive Disclosure | The skill references external policies and templates (e.g., policies/handoffs.md, templates/handoff.md, GLOBAL.md) which is good progressive disclosure, but the main body itself is a long flat list of sections without clear quick-start vs. advanced separation. The references are mentioned but not clearly signaled with navigation cues or descriptions of what each contains. | 2 / 3 |
Total | 7 / 12 Passed |
Validation
100%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.
d87ad31
Table of Contents
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.