CtrlK
BlogDocsLog inGet started
Tessl Logo

harness-plan-brief

Generate a Plan Brief HTML for non-engineer vibecoders before implementation starts. Searches harness-mem (project-only) for relevant past decisions, patterns, and Plans archive entries, then renders a single-file HTML artifact summarizing understanding, options, risks, acceptance criteria, and confidence. Use when the user requests a planning preview, a non-engineer-friendly summary before approval, or says: plan brief, planning preview, 計画概要, 計画レビュー. Do NOT load for: actual implementation, code review, release work.

75

Quality

94%

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

88%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 a well-sequenced, highly actionable workflow with explicit validation and failure-handling checkpoints, leaning on executable shell snippets. Its main weakness is progressive_disclosure: the body references multiple scripts and a schema file that are not actually present as bundle files, and a few meta-rationale asides add token weight without aiding execution.

Suggestions

Add the referenced scripts (plan-brief-compile.sh, render-html.sh, plan-brief-open.sh, plan-brief-record-decision.sh) and schemas/plan-brief-context.v1.schema.json to the bundle so progressive_disclosure references resolve to real files.

Trim the inline rationale blockquotes (e.g. the 'D43 判断 1 の根拠' and 'Phase 105.3 以降' readiness explanations) or move them to a references doc to improve conciseness.

Move the full plan-brief-context.v1 schema listing out of the body into the referenced schema file, keeping only a short pointer inline, to better separate overview from detail.

DimensionReasoningScore

Conciseness

Mostly efficient: action-oriented tables, inline code blocks, and tight step descriptions, but some meta-commentary and rationale paragraphs (e.g. the multi-line 'D43 判断 1 の根拠' blockquote and the 'Phase 105.3 以降...' readiness explanation) pad beyond what Claude needs; this is the 4 anchor (efficient, minor over-explanation that could be trimmed), not 5 because those asides do not all earn their tokens.

4 / 5

Actionability

Provides fully executable, copy-paste-ready commands throughout: 'PROJECT_NAME="$(basename "$(git rev-parse --show-toplevel)")"', the full mcp search parameter block, the jq -n context-build snippet, and 'bash scripts/render-html.sh --template plan-brief --data ...'; matches the 5 anchor with specific examples covering common cases; not a 4 because the concrete commands cover the full flow rather than having gaps.

5 / 5

Workflow Clarity

Clear six-step sequence (resolve project → search → assemble context → render HTML → open → approval) with explicit validation/fallback checkpoints in the 失敗時の挙動 table (mem search unreachable → continue with empty arrays; git fail → default; render fail → exit 1; open fail → best-effort exit 0) and per-step guardrails like 'strict_project: true を必ず指定'; matches the 5 anchor with explicit validation steps and error recovery feedback loops; not a 4 because checkpoints and recovery are explicit, not implicit.

5 / 5

Progressive Disclosure

Good structure with clear sections and well-signaled one-level-deep references to scripts (plan-brief-compile.sh, render-html.sh, plan-brief-open.sh, plan-brief-record-decision.sh) and the schema file, but no bundle files actually exist in references/scripts/assets, so the referenced scripts are not present as separate files — they are signaled but unverifiable, fitting the 4 anchor (good structure, mostly clear, minor organization gaps) rather than 5 (clear overview with appropriately split content in real referenced files).

4 / 5

Total

18

/

20

Passed

Description

100%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 exemplary: third-person voice, concrete actions, comprehensive natural trigger phrases (EN + JP), explicit 'Use when' guidance, and clear anti-triggers. It cleanly hits the top anchor on all four dimensions.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('Searches harness-mem', 'renders a single-file HTML artifact summarizing understanding, options, risks, acceptance criteria, and confidence') with comprehensive coverage, matching the score-5 anchor; it is not the 4 anchor because coverage of artifact contents is complete rather than having minor gaps.

5 / 5

Completeness

Explicitly answers both what ('Generate a Plan Brief HTML... renders a single-file HTML artifact...') and when ('Use when the user requests a planning preview... or says: plan brief, planning preview...') with concrete trigger phrases, matching the 5 anchor exactly.

5 / 5

Trigger Term Quality

Includes natural user phrases 'plan brief', 'planning preview', plus the Japanese synonyms '計画概要' and '計計画レビュー' and 'a non-engineer-friendly summary before approval', giving comprehensive synonym coverage as required for a 5; not a 4 because common variations are already present.

5 / 5

Distinctiveness Conflict Risk

Has a clear niche (pre-implementation non-engineer planning preview) and explicit anti-triggers ('Do NOT load for: actual implementation, code review, release work'), giving minimal conflict risk per the 5 anchor; not a 4 because the explicit exclusions remove even minor overlap risk.

5 / 5

Total

20

/

20

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing, 1 deeper-than-1-level

Warning

referenced_paths_exist

Referenced path issues: 11 missing

Warning

Total

14

/

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.