CtrlK
BlogDocsLog inGet started
Tessl Logo

trellis-meta

Understand and customize the local Trellis architecture inside a user project. Use when modifying .trellis plus platform hooks, settings, agents, skills, commands, prompts, workflows, the channel runtime (trellis channel), bundled runtime agents under .trellis/agents/, selectable workflow templates, registry-backed spec refresh, cross-session memory (trellis mem) generated by trellis init, or AI-facing bundled skills (trellis-channel, trellis-session-insight, trellis-spec-bootstrap) and bundled-skill auto-dispatch flow.

61

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 ./.agents/skills/trellis-meta/SKILL.md

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

SKILL.md
Quality
Evals
Security

Quality

Content

71%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 well-structured and highly actionable, with excellent progressive disclosure through a verified, one-level-deep reference tree and clear destructive-operation guardrails. Its primary weakness is redundancy — the same facts are restated across the intro, Current Rules, and Do Not sections.

Suggestions

De-duplicate the channel event-log path and the node_modules/global-npm warning: state each once (e.g., in Current Rules) and reference it from elsewhere to tighten token use.

Convert the "Do Not" guardrails into a brief pre-edit validation checklist (check template-hashes -> prefer .new sidecar -> verify knob exists) to give the destructive-customization workflow an explicit feedback loop.

Surface the unlisted reference groups (claude-code, core, how-to-modify, meta) in the References section or note they exist, so the navigation map matches the actual bundle.

DimensionReasoningScore

Conciseness

The body is information-dense and assumes Claude's competence (no basic-concept padding), but several facts are repeated across sections (the channel event-log path appears 3+ times, the node_modules/global-npm warning appears in the intro, Current Rules, and Do Not), which is more than the minor trimming the level-4 anchor allows.

3 / 5

Actionability

It supplies concrete, executable guidance throughout — specific file paths, real CLI commands ("trellis channel spawn --agent <name>", "trellis workflow --template <id>", "task.py create --parent <slug>"), and exact config keys — with only minor high-level hand-offs to reference files, matching the mostly-executable anchor.

4 / 5

Workflow Clarity

"How To Use" gives a clear numbered reading sequence and the "Do Not" section supplies explicit guardrails for destructive customization (check ".trellis/.template-hashes.json first", prefer ".new" sidecars, cross-check config before claiming a knob exists), but these are cautionary checkpoints rather than a formal validate-fix-retry loop.

4 / 5

Progressive Disclosure

The body is a true overview that delegates detail to a well-organized references tree grouped into Local Architecture, Platform Files, and Local Customization, each entry carrying a one-line description; references are one level deep, clearly signaled, and all referenced paths exist on disk.

5 / 5

Total

16

/

20

Passed

Description

73%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 clearly scopes a distinct Trellis-meta niche with an explicit and detailed Use-when clause and rich trigger coverage. Its main weakness is that the stated actions are limited to the generic verbs "understand" and "customize" rather than enumerating concrete operations.

Suggestions

Lead with 2-3 concrete verbs (e.g., "Inspect, modify, and re-route the local Trellis architecture") instead of the generic "Understand and customize" to lift specificity.

Trim internal jargon in the Use-when list ("registry-backed spec refresh", "bundled-skill auto-dispatch flow") in favor of terms a user would naturally say.

Add one or two natural synonyms or file extensions (e.g., ".trellis/config.yaml", "workflow template") to push trigger-term coverage toward comprehensive.

DimensionReasoningScore

Specificity

The description names the domain comprehensively ("the local Trellis architecture", ".trellis", platform hooks, agents, skills, channel runtime, spec refresh, cross-session memory) but offers only two generic verbs ("Understand and customize"), matching the anchor that names a domain with 1-2 concrete actions rather than several distinct actions.

3 / 5

Completeness

Both halves are explicitly present: a clear "what" ("Understand and customize the local Trellis architecture inside a user project") and a detailed "when" ("Use when modifying ..."), but the "what" verbs are broad, so it falls between the explicit-but-could-be-sharper anchor and the fully concrete anchor.

4 / 5

Trigger Term Quality

"Use when modifying .trellis plus platform hooks, settings, agents, skills, commands, prompts, workflows" supplies good natural keyword coverage a Trellis user would say, though it is diluted by internal jargon like "registry-backed spec refresh" and "bundled-skill auto-dispatch flow", so it sits just below the comprehensive-synonym anchor.

4 / 5

Distinctiveness Conflict Risk

It carves a clear, narrowly specific niche (Trellis meta-customization) with distinct triggers tied to named Trellis surfaces and bundled skills, giving minimal conflict risk with unrelated skills.

5 / 5

Total

16

/

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.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 27 deeper-than-1-level

Warning

Total

15

/

16

Passed

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.