CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dogfood

Walk this project's own personas through its real user-facing surface before a human does — a browser for a web app, the binary for a CLI, a client for an API or a tool server — and report the friction and breakage as candidate tickets. Use when a branch is functionally complete and somebody is about to say "ship it", or when the user asks to dogfood a branch, run an end-to-end pass, or test the changed flows as a user.

Quality

85%

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

81%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.

This is a well-crafted instruction skill: a clearly sequenced five-step procedure with explicit stop conditions and validation checkpoints, concrete report and ticket formats, and a genuinely useful trust-boundary section. The main improvable areas are trimming the motivational intro and provenance footer, and splitting the report/matrix format detail into a reference file.

DimensionReasoningScore

Conciseness

The body is dense with instructive content ("Record what you observed, not what you expected to observe", "A finding without evidence is an opinion") and assumes Claude's competence — it never explains concepts like browsers or APIs. Minor instances could be trimmed: the three-paragraph motivational intro and the provenance footer add tokens that do not guide execution, placing it just below the lean 5 anchor.

4 / 5

Actionability

Gives concrete, executable direction: "Diff the branch against the trunk it will merge into", "One row per (persona, journey, assertion) triple, written before anything is launched", a specified report location ("docs/dogfood-reports/, named for the branch and the date") with its required contents, and enumerated ticket fields, plus a worked positive/negative phrasing example ("upload a file, then share it with a colleague" — not "POST /files then PATCH /acl"). Not 5 because the mechanics of actually driving each surface (browser automation, tool-client invocation) are left unspecified, an acknowledged consequence of being product-agnostic.

4 / 5

Workflow Clarity

Five clearly sequenced steps with explicit validation checkpoints and error-recovery loops: stop if the product article is missing or unfilled ("Guessing a persona produces a report about an imaginary user"), stop on a red suite ("If the tests are failing, the branch is not ready to be dogfooded"), capture startup output for later diagnosis, and "If the surface will not come up, that is the finding, and the run stops here". The pre-written matrix functions as a checklist, matching the 5 anchor's pattern of explicit validation steps and feedback loops.

5 / 5

Progressive Disclosure

A well-sectioned single file whose external references ("constitution/local-product.md.template", "/to-tickets", the LICENSE provenance file) are one level deep and clearly signaled, with no bundle files present and no nested reference chains. Not 5 because at ~144 lines it exceeds the simple-skill exception, and detail like the report/ticket format and matrix construction (~30 lines) is kept inline where a reference file could carry it.

4 / 5

Total

17

/

20

Passed

Description

83%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: it states what the skill does in concrete, product-surface-specific terms and gives an explicit multi-trigger 'Use when' clause with natural phrasing a user would actually say. Its only weaknesses are a couple of missing trigger synonyms and slight overlap potential with adjacent e2e-testing skills.

DimensionReasoningScore

Specificity

Lists several concrete actions — "Walk this project's own personas through its real user-facing surface" and "report the friction and breakage as candidate tickets" — each grounded with surface instantiations ("a browser for a web app, the binary for a CLI, a client for an API or a tool server"). Falls short of the 5 anchor because the set of distinct actions is smaller and less comprehensive than the anchor's full coverage example, but exceeds 3 since the domain and multiple actions are named concretely.

4 / 5

Completeness

Explicitly answers both questions: the "what" ("Walk this project's own personas through its real user-facing surface... and report the friction and breakage as candidate tickets") and the "when" via a full "Use when" clause with concrete trigger phrases ("a branch is functionally complete", "dogfood a branch", "run an end-to-end pass", "test the changed flows as a user"). This matches the 5 anchor exactly; the when is fully explicit, so 4's weaker-when caveat does not apply.

5 / 5

Trigger Term Quality

Contains natural phrases users would actually say: "dogfood a branch", "run an end-to-end pass", "test the changed flows as a user", and "about to say 'ship it'". Good keyword coverage with a few natural terms missing (e.g., "smoke test", "QA pass", "try it as a user"), which places it at 4 rather than comprehensive synonym-level coverage.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (pre-merge dogfooding as declared personas) anchored by the distinctive term "dogfood" and persona/user framing, so it is mostly distinct with minimal generic conflict. Minor overlap risk remains with closely related e2e-testing/QA skills through the phrase "run an end-to-end pass", keeping it below the clear-niche, minimal-conflict 5 anchor.

4 / 5

Total

17

/

20

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

agranado2k/google-books-clojure

Path

.claude/skills/dogfood/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.