Name: breezing
Rating: 40 (1 reviews)
Author: Chachamaru127

breezing

Team execution mode — backward-compatible alias for harness-work with team orchestration. Composer/composer 2.5 maps to the cursor backend.

Quality

40%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./opencode/skills/breezing/SKILL.md

The canonical home for this skill is breezing in Chachamaru127/claude-code-harness

Quality

Content

51%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 content is highly actionable and the team-execution workflow is clearly sequenced with real review/validation gates. Its main weaknesses are verbosity (much restated harness-work content and repeated discipline reminders) and mixed progressive disclosure: two of four referenced files are missing from the bundle while other detail-heavy material is inlined.

Suggestions

Move the CC concurrency-limit and 1h-cache discussion into a reference file and keep only the one-line directive in SKILL.md.

Fix or remove the dangling references to references/backend-selection.md and references/completion-report.md, which are cited but not present in the bundle.

De-duplicate the repeated 'bin/harness work-mode on/off on every run boundary' instructions so the rule is stated once and referenced thereafter.

Dimension	Reasoning	Score
Conciseness	The ~480-line body is noticeably verbose: it restates harness-work material, repeats the 'bin/harness work-mode on/off' pairing across several sections, and explains CC concurrency limits and cache TTLs at length rather than pointing to one reference.	2 / 5
Actionability	It supplies concrete, copy-pasteable commands (resolver, model-routing.sh, cursor-companion.sh task invocations, stdin pattern) and specific flag semantics; minor gaps where behavior is described abstractly rather than as a command.	4 / 5
Workflow Clarity	The plan→work→review→report pipeline is explicitly sequenced with checkpoints (Plan gate, Integrated Review Gate before Phase C, APPROVE-or-iterate up to 3x with escalation), and review/batch operations carry validation; minor gaps from cross-references to harness-work for canonical steps.	4 / 5
Progressive Disclosure	Two referenced files (lean-path-detail.md, monitor-and-learning.md) exist and are one level deep, but the body also references backend-selection.md and completion-report.md which are NOT in the bundle; meanwhile large swaths that belong in references (concurrency limits, cache notes) are inlined.	3 / 5
	Total	13 / 20 Passed

Description

28%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 is essentially a self-referential alias note heavy on internal jargon ('harness-work', 'composer 2.5') and offers no concrete actions, no natural trigger terms, and no 'Use when...' guidance. It distinguishes itself poorly from the skill it aliases.

Suggestions

Add a concrete 'Use when...' clause naming the natural user phrasing that should invoke this skill (e.g. 'run the whole plan→implement→review→report pipeline in one command').

Replace internal jargon ('composer/composer 2.5 maps to the cursor backend') with capability-oriented language describing what the user gets.

State 2-3 concrete actions the skill performs (orchestrates Lead/Worker/Reviewer, runs integrated review gate, auto-updates Plans.md) instead of only 'alias for harness-work'.

Dimension	Reasoning	Score
Specificity	Names the domain ('team execution mode', 'harness-work with team orchestration') but the only concrete action stated is being a 'backward-compatible alias' that 'maps to the cursor backend' — no concrete capabilities beyond delegating.	2 / 5
Completeness	Has a vague 'what' (alias for harness-work with team orchestration) but no 'when' / 'Use when...' clause at all, so per the judging guideline completeness is capped at 3 and lands lower given the vague what.	2 / 5
Trigger Term Quality	Terms like 'harness-work', 'team orchestration', 'composer/composer 2.5' are internal jargon, not the natural phrases a user would say when needing this skill; no common synonyms or natural trigger language.	2 / 5
Distinctiveness Conflict Risk	It is somewhat specific (a team execution alias) but its stated purpose overlaps heavily with harness-work itself, so a user could easily trigger the wrong skill; minor-to-moderate conflict risk.	3 / 5
	Total	9 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 7 missing, 7 deeper-than-1-level, 1 suspicious	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: Chachamaru127/claude-code-harness
Path: opencode/skills/breezing/SKILL.md
Commit: 2b2b748

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.