CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

openspec

Spec-driven development with OpenSpec. Use when the user mentions OpenSpec, OPSX, spec-driven development, SDD, change proposals, delta specs, /opsx commands, or wants structured planning for features, migrations, refactors, or bug fixes.

Quality

83%

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

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 well-structured, largely executable instruction skill: clear sequencing, explicit validation requirements, and a real externalized installer script. The main gaps are the absence of an explicit validate-fail-retry feedback loop and isolated prose redundancy in the NVM section.

Suggestions

Add an explicit feedback loop after validation (e.g., 'If openspec validate fails, fix the cited artifacts and re-run before proceeding to the next task') to raise workflow_clarity.

Tighten the CLI Runtime Resolution section: the 'Do not conclude...' paragraph largely restates the resolution sequence, so trim it to the one non-obvious signal (~/.agents/bin/openspec reporting 'env: node: No such file or directory').

Reference the installer with its bundle-relative path (scripts/install-openspec) in addition to the shared ~/.agents path so the bundle structure is self-evident.

DimensionReasoningScore

Conciseness

The body is dense with tool-specific guidance (NVM resolution sequence, installer placement policy, OPSX workflow) and avoids explaining concepts Claude already knows, but the NVM section restates its own code twice — 'Do not conclude that OpenSpec is unavailable merely because it is absent from the initial PATH' repeats what the resolution sequence already demonstrates — and the symlink path explanation appears in two sections. Not level 5 because every token does not earn its place; not level 3 because the padding is isolated to a few sentences rather than whole sections.

4 / 5

Actionability

Provides copy-paste bash for NVM resolution, installer invocation, and CLI state/validation commands ('openspec list', 'openspec validate --all'), plus a fallback ('perform the same workflow directly in files under openspec/changes/<change>/') for tools without slash commands. Not level 5 because several workflow steps remain directive rather than executable ('Explore unclear requirements before creating artifacts', 'Review and refine proposal.md, delta specs, design.md, and tasks.md') with no examples of well-formed artifacts.

4 / 5

Workflow Clarity

The Default Workflow is a clear six-step sequence (explore, propose, review, apply tasks one at a time, verify against proposal/specs/design/tasks, sync and archive) with explicit validation mandated in Artifact Standards ('Run OpenSpec validation and the repository's canonical validation before claiming completion'). Not level 5 because there is no error-recovery feedback loop (nothing says what to do when validation fails or a task check reveals a defect); not level 3 because checkpoints are explicit rather than implicit.

4 / 5

Progressive Disclosure

The single bundle file (scripts/install-openspec) is appropriately externalized and the ~107-line body is organized into clear, well-scoped sections with no content that clearly belongs in a separate file. Not level 5 because the installer is referenced via the shared '~/.agents/skills/openspec/scripts/install-openspec' path rather than a clearly signaled relative bundle path, and the machine-specific NVM detail could live in a reference file.

4 / 5

Total

16

/

20

Passed

Description

86%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.

A strong description with exemplary trigger coverage (product name, alias, methodology, abbreviation, commands, and intent) and an explicit, well-scoped 'Use when' clause. The only weaknesses are a thin action enumeration in the 'what' portion and a slightly broad intent-based trigger that could overlap with generic planning skills.

DimensionReasoningScore

Specificity

The description names the domain ('Spec-driven development with OpenSpec') and one or two actions ('structured planning'), which matches the anchor for a named domain with 1-2 concrete actions. Not level 4 because it never lists several specific actions the skill performs (proposing changes, generating delta specs, applying task lists, archiving) — 'change proposals' and 'delta specs' appear only as trigger nouns; not level 2 because it is tied to a specific tool and methodology rather than generic language.

3 / 5

Completeness

Both questions are answered explicitly: 'what' via 'Spec-driven development with OpenSpec' and 'when' via a fully enumerated 'Use when the user mentions...' clause with concrete trigger phrases. Not level 4 because the 'when' is already maximally explicit and specific, so the level-4 anchor ('when could be more explicit or specific') does not fit.

5 / 5

Trigger Term Quality

Coverage is comprehensive with synonyms and abbreviations: 'OpenSpec', 'OPSX', 'spec-driven development', 'SDD', plus artifact-level terms ('change proposals', 'delta specs', '/opsx commands') and intent phrases ('structured planning for features, migrations, refactors, or bug fixes'). Not level 4 because no natural term a user would say for this skill is missing — the product name, its alias, the methodology, and its abbreviation are all present.

5 / 5

Distinctiveness Conflict Risk

The tool-specific triggers (OpenSpec, OPSX, SDD, delta specs) carve out a clear niche with minimal conflict risk. Not level 5 because the intent clause 'or wants structured planning for features, migrations, refactors, or bug fixes' is broad enough to fire on generic planning requests that overlap with general planning or spec-writing skills; not level 3 because the primary triggers are unambiguously tied to one tool.

4 / 5

Total

17

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

saski/arnesto

Path

.agents/skills/openspec/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.