CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

spec-driven-core

Processo universal e repetível para criar especificações a partir de qualquer input (texto, docs, código). Usado em Plan mode.

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/workflow/spec-driven-core/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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.

A lean, well-structured process skill: ordered phases with named deliverables, hard rules, and completion criteria make it genuinely actionable without padding. The main gaps are the absence of worked examples and an explicit error-recovery loop from the validation phase.

Suggestions

Add one compact worked example per critical phase — a well-formed observable requirement and a matching Given/When/Then acceptance criterion — to make the guidance copy-paste ready.

Add an explicit feedback loop after Phase 6: e.g., 'If validation fails, mark the gap as OQ-xxx and return to the phase that produced the artifact' — this closes the workflow-clarity gap.

Move per-phase detail (requirement rules, ADR format, glossary template) into one-level-deep reference files to keep SKILL.md a lean overview.

DimensionReasoningScore

Conciseness

The body is terse throughout — each phase is 2-4 lines with explicit outputs, and there is no explanation of concepts Claude already knows. Only minor meta-statements could be trimmed ("This skill is educational and operational", "never overwhelm; proceed incrementally"), which fits 'efficient; minor instances of over-explanation that could be trimmed' rather than the fully lean anchor.

4 / 5

Actionability

Concrete, executable guidance for an instruction-only skill: exact artifacts per phase (system-context.md, prd.md, acceptance-criteria.md), explicit questions to answer, hard rules ('each requirement must be observable; no implementation details'), the Given/When/Then convention, the OQ-xxx convention, and a concrete output directory tree. It falls short of fully copy-paste-ready because there is no worked example of a well-formed requirement or acceptance criterion.

4 / 5

Workflow Clarity

The sequence is explicit and ordered (Phase 0 through Phase 6) with defined outputs per phase, a dedicated validation phase ('Verify traceability, completeness, no hidden assumptions, no vague language'), and completion criteria as a final checklist. It misses the 5 anchor because there is no feedback loop specifying what to revisit when Phase 6 validation fails.

4 / 5

Progressive Disclosure

No bundle files exist, and the ~95-line body is well-sectioned and self-contained with a clear 'Route to Auxiliary Skills' navigation section. It scores 4 rather than 5 because it exceeds the simple-skill threshold (under 50 lines) and per-phase detail (e.g., requirement-writing rules, ADR format) could live in one-level-deep reference files.

4 / 5

Total

16

/

20

Passed

Description

58%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 states a clear what and an explicit when (Plan mode) in third person, but is thin on concrete capabilities and natural trigger synonyms — the richer trigger terms live only in the separate triggers field. The 'universal / any input' framing also raises conflict risk with related spec and planning skills.

Suggestions

Enumerate the concrete actions in the description itself, e.g., 'Cria contexto do sistema, requisitos funcionais, critérios de aceitação e ADRs a partir de texto, docs ou código' — this lifts specificity from one generic action to several concrete ones.

Move natural trigger phrases into the description, e.g., 'Use quando o usuário pedir specs, requisitos, um PRD, ou entrar em Plan mode', so the description alone covers the synonyms currently isolated in the triggers field.

Narrow the 'universal / qualquer input' claim to the actual niche (specification writing) to reduce overlap with sibling skills like requirements-extractor and spec-normalizer.

DimensionReasoningScore

Specificity

"criar especificações a partir de qualquer input (texto, docs, código)" names the domain and one concrete action with input types, matching 'names domain and 1-2 concrete actions, but not comprehensive'. It does not list several specific actions (e.g., what artifacts the process produces), so it falls short of 4.

3 / 5

Completeness

Both what ("Processo universal e repetível para criar especificações...") and when ("Usado em Plan mode") are explicitly stated, matching 'has both what and when; when could be more explicit or specific'. It is not 5 because the when lacks concrete trigger phrases covering the common cases (user asking for specs, requirements, or a PRD).

4 / 5

Trigger Term Quality

The description's keywords are limited to "especificações", "Plan mode", and "texto, docs, código"; natural synonyms like spec, requirements, PRD, and planning appear only in the separate triggers field, not the description itself. This matches 'some relevant keywords but missing common variations or synonyms'.

3 / 5

Distinctiveness Conflict Risk

Spec-creation is a recognizable niche, but the "universal" / "qualquer input" claim plus broad planning triggers create real overlap with requirements-gathering and planning skills — the body itself routes to five sibling spec skills (requirements-extractor, spec-normalizer, etc.). This fits 'somewhat specific but could still overlap with similar skills' better than the mostly-distinct anchor.

3 / 5

Total

13

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

fernandezbaptiste/cursor-multiagent-system

Path

skills/workflow/spec-driven-core/SKILL.md

Reviewed

Table of Contents

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.