CtrlK
BlogDocsLog inGet started
Tessl Logo

trellis-channel

Use Trellis channel for live multi-agent collaboration, spawned workers, cross-agent review, progress inspection, forum channels, and channel log debugging.

64

Quality

81%

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

78%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-built index-style skill: lean body, executable first commands, an explicit inspect-before-asking checkpoint, and an unambiguous intent-based routing table to real, one-level-deep reference files. Remaining gains are minor: deduplicate the user-signals list against the routing table and surface one inline example for the most common operation (spawn/send).

DimensionReasoningScore

Conciseness

The body is lean for an index skill: it assumes competence, defers detail to reference files, and carries only non-obvious domain knowledge (e.g., the dispatcher wait-pattern failure mode). Minor trimming is possible — the 'Typical user signals' list partially duplicates the phrases in the Route By User Intent table, and the two-sentence index preamble could be one — so it fits anchor 4 rather than 5.

4 / 5

Actionability

First Commands are copy-paste-ready ('trellis channel list --scope global --all', 'trellis channel forum <board> --scope global') and Core Rules give concrete flag-level guidance (--type forum, --raw, --kind done). Most common operations are executable inline; a few gaps remain (no inline example of spawning a worker or sending a message, though these are deliberately deferred to reference files), placing it between anchors 4 and 5, closer to 4.

4 / 5

Workflow Clarity

A clear sequence is present: run first commands → inspect the named channel/thread before asking for background (an explicit checkpoint) → route by intent table → follow core rules. Checkpoints are mostly present, though verification of outcomes (e.g., confirming a spawned worker actually started, or checking forum state after posting) is left implicit in the reference files, keeping it below anchor 5.

4 / 5

Progressive Disclosure

The body explicitly declares 'This skill is an index. Load only the reference file for the current job', provides an intent→file routing table, and lists all five reference files with one-line descriptions. All referenced files exist (workflows.md's patterns A–F and command-reference.md's 'tag-vs-kind' section were verified present), references are one level deep, and the 'Not For' section adds boundary navigation — a clear match for anchor 5.

5 / 5

Total

17

/

20

Passed

Description

75%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 solid, concise description that names the tool and enumerates several specific capability areas with mostly natural trigger terms. Its main weakness is the absence of an explicit 'Use when...' trigger clause and of natural synonyms (brainstorm, discuss, stalled) that would sharpen routing.

Suggestions

Add an explicit 'Use when...' clause naming user situations, e.g., 'Use when the user wants agents to brainstorm/discuss, spawn an implement/check worker, cross-review work, capture issues on a forum board, or debug a stalled worker or truncated progress.'

Include natural synonyms users actually say — 'brainstorm', 'discuss with another agent', 'worker stalled / no output', 'thread' — to round out trigger term coverage.

DimensionReasoningScore

Specificity

Lists several specific capability areas — 'spawned workers', 'cross-agent review', 'progress inspection', 'forum channels', 'channel log debugging' — but they are noun phrases rather than concrete verb-actions, and coverage gaps exist (e.g., thread/context management, brainstorm workflows). This sits above anchor 3 (only 1-2 concrete actions) but the nominal phrasing keeps it below anchor 5's comprehensive concrete-action coverage.

4 / 5

Completeness

The 'what' is clear (Trellis channel runtime for collaboration, workers, review, progress, forums, debugging) and the 'for X, Y, Z' construction provides equivalent trigger guidance, but the 'when' is embedded in the capability list rather than stated as an explicit 'Use when...' clause. This matches anchor 4 ('when' could be more explicit) — the capability-list phrasing functions as trigger guidance, so it does not fall to the anchor-3 cap.

4 / 5

Trigger Term Quality

Good natural keywords users would say: 'spawned workers', 'review', 'forum channels', 'progress', 'debugging', 'multi-agent collaboration'. A few natural terms the body itself surfaces are missing from the description — 'brainstorm', 'discuss', 'worker stalled', 'thread' — so it falls just short of anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

'Trellis channel' is a named, niche tool with tool-specific triggers (forum channels, channel log debugging, spawned workers), giving it a clear niche. Minor overlap risk remains with sibling trellis skills (e.g., trellis-session-insight / memory search referenced in the body), which prevents anchor 5.

4 / 5

Total

16

/

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.