CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

prototype

Answer a design or feasibility question with explicitly throwaway spike code — never committed, findings recorded in the diary or an ADR draft. Use when a decision is blocked on "would X even work?" (library capability, API behavior, performance shape). Not for building features — that's /tdd via /implement.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

100%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.

An exemplary lean skill body: dense rule statements with no filler, a copy-paste-ready trace command, and a procedure with genuine validation checkpoints (docs-first check, record-before-delete, timebox stop condition). Organization suits its size perfectly — no bundle files are needed and none are missing.

DimensionReasoningScore

Conciseness

The ~28-line body is dense and every line carries non-obvious policy (scratch-dir placement, suspension of the test-first rule, trace no-op behavior, version-citation requirement) with zero explanations of concepts Claude already knows. The rule-4 parenthetical justifies a governance exception rather than padding, so nothing is trimmable as the 4 anchor would require.

5 / 5

Actionability

Provides a copy-paste-ready command ('sh scripts/trace.sh emit kind=spike.verdict [subject=...] outcome=true|false|inconclusive --blob <file> ... || :'), concrete destinations ('docs/diary.md', session scratchpad or $TMPDIR), and a worked falsifiable-question example — fully executable guidance for an instruction-only skill.

5 / 5

Workflow Clarity

The procedure is clearly sequenced with explicit checkpoints: docs-first check (step 1), evidence capture (step 3), record-before-delete ordering protecting the one destructive operation (step 4), and a timebox feedback loop ('A spike that runs long is answering a different, bigger question — stop and say so') plus an inconclusive outcome path. This matches the 5 anchor's explicit-validation-and-recovery profile; the destructive delete is validated by recording the finding first, so the score-3 cap does not apply.

5 / 5

Progressive Disclosure

The skill is under 50 lines, needs no external reference files (none exist in the bundle), and is cleanly organized into a title, orienting intro, Rules, and Procedure sections — meeting the rubric's simple-skill criterion for a 5. External paths mentioned (constitution/local-engineering.md, docs/diary.md, scripts/trace.sh) are project files, not skill-bundle references, and the trace call degrades to a silent no-op when unconfigured.

5 / 5

Total

20

/

20

Passed

Description

87%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 that explicitly covers what, when, and what-not, with a concrete trigger phrase and explicit disambiguation from sibling skills. The only weakness is moderate rather than comprehensive keyword coverage — natural synonyms like 'proof of concept' or 'experiment' are absent.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'Answer a design or feasibility question with explicitly throwaway spike code', 'never committed', 'findings recorded in the diary or an ADR draft' — but omits mechanics the body treats as central (timebox, scratch-dir placement), leaving minor coverage gaps rather than the comprehensive coverage of the 5 anchor.

4 / 5

Completeness

Clearly answers both what ('Answer a design or feasibility question with explicitly throwaway spike code — never committed, findings recorded in the diary or an ADR draft') and when ('Use when a decision is blocked on "would X even work?"') with a concrete trigger phrase, plus an explicit negative boundary ('Not for building features — that's /tdd via /implement'), matching the 5 anchor exactly.

5 / 5

Trigger Term Quality

Includes natural phrases users would actually say — "would X even work?", 'feasibility question', 'spike code', 'library capability, API behavior, performance shape' — but misses common synonyms such as 'proof of concept', 'POC', or 'experiment', which keeps it below the comprehensive-synonym coverage of the 5 anchor.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (throwaway feasibility spikes) with distinct triggers and explicit disambiguation against adjacent skills ('Not for building features — that's /tdd via /implement'), so conflict risk is minimal — the 5 anchor's best fit.

5 / 5

Total

18

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

referenced_paths_exist

Referenced path issues: 1 possibly missing (scripts/trace.sh)

Warning

Total

13

/

16

Passed

Repository

agranado2k/agentic-sdlc

Path

.agents/skills/prototype/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.