CtrlK
BlogDocsLog inGet started
Tessl Logo

trellis-update-spec

Captures executable contracts and coding conventions into .trellis/spec/ documents. Use when learning something valuable from debugging, implementing, or discussion that should be preserved for future sessions.

53

Quality

67%

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 ./.claude/skills/trellis-update-spec/SKILL.md

The canonical home for this skill is trellis-update-spec in mindfold-ai/Trellis

SKILL.md
Quality
Evals
Security

Quality

Content

56%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.

The body is highly actionable with concrete templates and a clear, checklist-validated workflow, but it is noticeably verbose — classification guidance is triplicated across three sections and several prose sections pad without instructing. With no bundle files, the long template library is inlined where a one-level reference file would keep the SKILL.md lean.

Suggestions

Merge the classification guidance into one place: keep the Step 2 classify table and delete the overlapping "When to Update" trigger table and the Interactive Mode question lists.

Move the ~140-line template library to references/templates.md and keep one-line pointers in SKILL.md, keeping only the mandatory infra/cross-layer template inline.

Cut the "Core Philosophy" and "Relationship to Other Commands" sections, or reduce them to 2-3 lines, since they repeat motivations already stated in the opening and add no executable guidance.

DimensionReasoningScore

Conciseness

Classification guidance is repeated three times (the "When to Update" table, Step 2's classify table, and the Interactive Mode questions), the 7-section list appears in both "Mandatory Output" and the template, and "Core Philosophy", "Relationship to Other Commands", and "Key Insight" prose pad without instructing — several padded sections, though the templates themselves earn their place (anchor 2, not 1 or 3).

2 / 5

Actionability

Provides copy-paste-ready markdown templates with exact paths (.trellis/spec/<layer>/*.md), a bash command, and exact section skeletons; the gap is that no filled-in concrete example of the mandatory 7-section template is shown (anchor 4, not 5).

4 / 5

Workflow Clarity

A clear 5-step sequence (identify, classify, read target, update, update index) with an explicit Quality Checklist as a final verification checkpoint; minor gap is no post-update verification step such as re-reading for duplication (anchor 4, not 5).

4 / 5

Progressive Disclosure

No bundle files exist and the entire ~350-line skill is inline; the ~140-line templates section is a clear candidate for a references/templates.md file, though the SKILL.md itself is well-organized with clear headers (anchor 3, not 2 or 4).

3 / 5

Total

13

/

20

Passed

Description

62%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 answers both what and when in third person with a distinct niche, but lists only one capability and relies on somewhat abstract trigger language ("learning something valuable") rather than concrete user-sayable phrases. It is solid but leaves room for more specific action coverage and richer trigger synonyms.

Suggestions

List 2-3 concrete actions instead of one, e.g. "Captures executable contracts, signatures, and validation/error matrices into .trellis/spec/ documents".

Add natural trigger phrases users would say, such as "Use when the user says 'remember this', 'document this convention', or 'we learned X the hard way', or after fixing a subtle bug".

DimensionReasoningScore

Specificity

"Captures executable contracts and coding conventions into .trellis/spec/ documents" names the domain, concrete objects, and an exact destination path, but lists only one action rather than several specific actions, so coverage is not comprehensive (anchor 3, not 4).

3 / 5

Completeness

Has both a clear "what" (captures contracts/conventions into .trellis/spec/) and an explicit "Use when..." clause, but the trigger "learning something valuable" is abstract rather than the concrete trigger phrases of anchor 5.

4 / 5

Trigger Term Quality

Includes relevant natural keywords ("debugging", "implementing", "discussion", "learning") but misses common variations users would actually say such as "lessons learned", "document this", or "write a spec" (anchor 3, not 4).

3 / 5

Distinctiveness Conflict Risk

The .trellis/spec/ destination gives it a distinct niche with minimal confusion, but broad triggers like "debugging", "implementing", and "discussion" create minor overlap risk with general documentation skills, so it does not fully reach anchor 5.

4 / 5

Total

14

/

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
mindfold-ai/Trellis
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.