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.

76

Quality

96%

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

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

SKILL.md
Quality
Evals
Security

Quality

Content

92%

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 tight, highly actionable execution spec with clear sequencing and failure handling. Its only weakness is progressive disclosure: it points to many referenced files that are not present in the bundle.

Suggestions

Ship the referenced bundle files (scripts/plan-brief-compile.sh, scripts/render-html.sh, scripts/plan-brief-open.sh, templates/html/plan-brief.html.template, schemas/plan-brief-context.v1.schema.json) so the in-body references resolve to real materials.

Inline a one-line fallback note for each referenced script describing its expected input/output, so the skill remains usable when a script is missing.

Add a short bundle-inventory section near the top listing which referenced files are required vs. optional, making the dependency on the bundle explicit.

DimensionReasoningScore

Conciseness

Lean and information-dense: tables for responsibility boundaries, outputs, and failure modes, plus exact bash/jq snippets, with no lecturing on concepts Claude already knows.

3 / 3

Actionability

Copy-paste ready: exact `git rev-parse --show-toplevel`/`basename` commands, precise MCP params with required fields flagged, a full jq template, and a concrete render-html.sh invocation.

3 / 3

Workflow Clarity

Clear numbered Step 1→6 sequence with an explicit 失敗時の挙動 table mapping each failure to recovery behavior, and a clearly separated cross-project alt path.

3 / 3

Progressive Disclosure

References are well-signaled and one-level-deep, but the referenced bundle files (scripts/*.sh, templates, schemas, tests) do not exist in the bundle, so navigation points to absent materials.

2 / 3

Total

11

/

12

Passed

Description

100%

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, well-triggered, complete, and distinctive, using third-person voice throughout. It cleanly separates what the skill does from when to invoke it and explicitly excludes overlapping use cases.

DimensionReasoningScore

Specificity

Lists multiple concrete actions: 'Searches harness-mem (project-only)', 'renders a single-file HTML artifact summarizing understanding, options, risks, acceptance criteria, and confidence' — concrete, not vague.

3 / 3

Completeness

Explicitly answers both what (search + render HTML summary) and when ('Use when the user requests a planning preview...'), plus negative triggers ('Do NOT load for...').

3 / 3

Trigger Term Quality

Good natural coverage users would say: 'planning preview', 'non-engineer-friendly summary before approval', and literal phrases 'plan brief, planning preview, 計画概要, 計画レビュー'.

3 / 3

Distinctiveness Conflict Risk

Clear niche (Plan Brief HTML for non-engineers) with distinct triggers and explicit exclusions (implementation, code review, release work), making wrong-skill triggering unlikely.

3 / 3

Total

12

/

12

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.