Name: harness-plan
Rating: 63.2 (1 reviews)
Author: Chachamaru127

harness-plan

HAR: Research-backed, team-validated task planning, Plans.md management, progress sync. Trigger: create a plan, add tasks, update Plans.md, mark complete, check progress. Do NOT load for: implementation, review, release.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./opencode/skills/harness-plan/SKILL.md

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

Quality

Content

63%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 dense, well-structured planning protocol with concrete subcommand syntax and validation-gated workflows, undermined by repeated policy blocks and broken references to scripts/docs paths absent from the bundle. Tightening redundancy and reconciling missing referenced paths would lift conciseness and progressive disclosure.

Suggestions

Consolidate the repeated 'Spec delta / Spec skip reason' and non-trivial planning-gate rules into a single section or defer fully to references/planning-quality.md to reduce conciseness padding.

Reconcile references to missing bundle paths (scripts/claude-longrun.sh, scripts/plans-issue-bridge.sh, scripts/plan-registry.sh, docs/plans/*, templates/schemas/plan-preapproval.v2.json) — either add the files or mark them as runtime/consumer-repo paths so navigation matches the actual bundle.

Make implicit workflow checkpoints explicit (e.g., add the literal validate→fix→retry loop for the pre-approval and DoD checks) to push workflow clarity toward a 5.

Dimension	Reasoning	Score
Conciseness	Mostly efficient domain-specific protocol rather than concepts Claude already knows, but the 'Spec delta / Spec skip reason' rules and non-trivial planning gate are restated across multiple sections and overlap with references/planning-quality.md, so it could be tightened.	3 / 5
Actionability	Concrete command syntax ('/harness-plan add タスク名: 詳細説明 [--phase フェーズ番号]'), marker tables, and a fixed-format 事前確認 block give mostly executable guidance with only minor gaps.	4 / 5
Workflow Clarity	The create flow is an explicit 11-step sequence with embedded validation checkpoints (品質チェック, 二正本チェック) and the pre-approval section adds a destructive/external-send feedback gate, though some checkpoints remain implicit rather than scripted.	4 / 5
Progressive Disclosure	Good one-level-deep signaling to real reference files (create.md, sync.md, planning-quality.md, criteria-design.md), but several referenced paths (scripts/.sh, docs/plans/.md, templates/schemas/plan-preapproval.v2.json) do not exist in the actual bundle, breaking navigation against the file listing.	3 / 5
	Total	14 / 20 Passed

Description

87%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 tight, third-person description that answers both what and when, supplies natural trigger phrases, and carves out a clear niche with an explicit exclusion boundary. Minor abstractness in 'Plans.md management'/'progress sync' and missing synonyms keep specificity and trigger quality just below perfect.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('create a plan, add tasks, update Plans.md, mark complete, check progress') alongside the planning domain, though 'Plans.md management' and 'progress sync' stay somewhat abstract, leaving minor coverage gaps that keep it below 5.	4 / 5
Completeness	An explicit 'Trigger:' clause supplies concrete 'when' phrases while the opening clause states the 'what' ('Research-backed, team-validated task planning, Plans.md management, progress sync'), clearly answering both questions.	5 / 5
Trigger Term Quality	Natural user phrases like 'create a plan', 'add tasks', 'mark complete', and 'check progress' give good keyword coverage with explicit exclusions, but it lacks common synonyms/variations that would push it to comprehensive.	4 / 5
Distinctiveness Conflict Risk	The narrow Plans.md planning niche plus an explicit 'Do NOT load for: implementation, review, release' boundary makes it clearly distinguishable with minimal conflict risk against sibling skills.	5 / 5
	Total	18 / 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: 6 missing, 6 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 8 missing	Warning

	Total	14 / 16 Passed

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

Reviewed: 5 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.