CtrlK
BlogDocsLog inGet started
Tessl Logo

harness-sync

HAR: Sync Plans.md with implementation. Drift detect, marker update, retrospective. Trigger: sync-status, where am I, check progress. --snapshot for snapshots. Do NOT load for: planning, implementation, review, release.

64

Quality

76%

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 ./opencode/skills/harness-sync/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 a well-structured, mostly lean workflow with concrete commands, sequenced steps, and an explicit validation gate plus confirmation prompt — appropriate for a drift-detection skill that mutates Plans.md. It is not a 5 primarily because some drift-detection checks are phrased as comparison descriptions rather than runnable commands, and a couple of sections (snapshot design notes, duplicated option tables) could be tightened.

Suggestions

Convert the Step 1.5 and Step 2 detection tables from descriptive comparisons ('コミット履歴 vs マーカー') into concrete shell snippets or jq/grep commands so each check is directly executable.

Tighten the duplicated option information: the Quick Reference table and the オプション table overlap; merge them or have the Quick Reference link to オプション.

Add a short validation/verification note in Step 4.5 (snapshot) and Step R4 (harness-mem write) confirming the write succeeded before proceeding.

DimensionReasoningScore

Conciseness

The body is largely lean and assumes Claude's competence — it uses tables and short code blocks to convey checks and commands rather than explaining what git or Plans.md are — with only minor spots (e.g., the snapshot design-intent blockquote and repeated Quick Reference / オプション tables) that could be trimmed.

4 / 5

Actionability

It provides concrete, executable commands (cat Plans.md, git status, git log, grep, jq pipelines, mkdir snapshot dir) and concrete output templates, but several detection steps are described as table rows ('コミット履歴 vs マーカー') rather than runnable commands, leaving minor gaps.

4 / 5

Workflow Clarity

Steps are clearly sequenced (Step 0 validation → Step 1 collection → Step 1.5 trace analysis → Step 2 drift → Step 3 update proposal → Step 4 summary → Step 5 next action → Step 6 retro) with a validation gate in Step 0 and a confirmation prompt in Step 3; minor validation gaps (e.g., no explicit feedback loop before writing the snapshot or recording to harness-mem) keep it just below the top anchor.

4 / 5

Progressive Disclosure

Well-organized into clearly headed sections covering the whole workflow at one level with no nested references and no separate bundle files to navigate; the content is self-contained and easy to scan, with only minor organization gaps (e.g., the Quick Reference and オプション tables overlap).

4 / 5

Total

16

/

20

Passed

Description

78%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 specific, includes natural trigger phrases, and cleanly separates this skill's niche from related harness skills via both positive triggers and negative-boundary guidance. Its main weakness is the terse, compressed 'Trigger:'/'Do NOT load for:' shorthand rather than a fluent 'Use when...' sentence, which keeps completeness just below the top anchor.

Suggestions

Convert the compressed 'Trigger:' and 'Do NOT load for:' shorthand into a full sentence such as 'Use when the user asks to check progress, sync Plans.md with implementation, or detect drift; do not load for planning, implementation, review, or release.'

Add one or two natural synonyms users might say (e.g. 'how far along', 'what's left') to broaden trigger coverage.

Spell out what 'marker update' updates (the cc:WIP/cc:完了 status markers) so the action is concrete rather than implicit.

DimensionReasoningScore

Specificity

Names the domain (Plans.md sync) and several concrete actions — 'Drift detect, marker update, retrospective' — plus flag-specific behavior ('--snapshot for snapshots'), showing specific capabilities with minor gaps in coverage of the underlying operations.

4 / 5

Completeness

Explicitly answers 'what' (sync Plans.md with implementation, drift detect, marker update, retrospective) and 'when' via 'Trigger:' clauses plus a 'Do NOT load for' negative-boundary clause; the trigger guidance is concrete but terse rather than a full 'Use when...' sentence.

4 / 5

Trigger Term Quality

Includes natural user phrases ('sync-status', 'where am I', 'check progress', '進捗確認') that users would plausibly say, with good coverage but a few common synonyms missing.

4 / 5

Distinctiveness Conflict Risk

A clearly defined niche (Plans.md ↔ implementation drift sync) with explicit negative boundaries ('Do NOT load for: planning, implementation, review, release'), minimizing overlap with sibling harness-* skills.

5 / 5

Total

17

/

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: 1 missing

Warning

Total

15

/

16

Passed

Repository
Chachamaru127/claude-code-harness
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.