sightline

Handle the parts of a plan that can't be decided by talking — answer every question the PRD, prompt, or codebase can already settle, then resolve what's genuinely unknowable with timeboxed probes and manipulable prototypes, gating progress on the user actually understanding what was decided. Keeps its map and findings on disk so an effort can span many context resets. Use this alongside or inside an interview-based planning skill such as plan-with-docs, whenever the user answers "I don't know yet", says a plan feels premature, says they need to prototype first, says they don't understand what the agent produced, or when the work is too big for one session. Also use it when a previous plan turned out to be wrong once building started, or to resume a fogged effort already underway.

Quality

84%

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

Quality

Content

73%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 terse, actionable, and has an exceptionally clear validated workflow. Its main weakness is progressive disclosure: the spine repeatedly points to cookbook files and a justfile that are not present in the bundle.

Suggestions

Add the missing cookbooks/ files referenced in the body (resolve-row.md, new-effort.md, land.md, board.md) or remove/inline those references so navigation targets resolve.

Provide the referenced justfile (or replace the `alias sight='just --justfile …'` instruction with a direct node invocation) so the documented commands actually run.

List the real scripts/ bundle (board-add, board-serve, lib, resume, spike) explicitly in a short index so the actual bundle structure is discoverable alongside the body's prose.

Dimension	Reasoning	Score
Conciseness	Deliberately terse ("Terse. Show, don't describe", "Under ~60 words of prose per turn") and assumes Claude's competence without re-explaining basics; a few sections run long but mostly earn their place, fitting efficient-but-trimmable rather than fully lean.	4 / 5
Actionability	Provides concrete executable commands (sight resume, sight spike, node scripts/resume.mjs), an exact disk layout, and a numbered turn loop; some guidance stays at the principle level rather than literal code, but that is appropriate for a recipe-driven instruction skill with minor gaps.	4 / 5
Workflow Clarity	The turn loop is an explicit ordered sequence with validation checkpoints (burn "refuses while findings/07*.md is missing", land-before-burn ordering, MOVED SINCE re-check) and feedback loops for error recovery, matching the explicit-validation anchor.	5 / 5
Progressive Disclosure	The spine is well organized and signals one-level-deep references (cookbooks/resolve-row.md, new-effort.md, land.md, board.md) plus a justfile, but none of those referenced files exist in the bundle — only scripts/*.mjs are present — so the navigation targets are dead and structure is effectively minimal/buried.	2 / 5
	Total	15 / 20 Passed

Description

95%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 strong, specific description with excellent trigger coverage and a clearly delineated niche. It answers both what and when with concrete, natural phrasing a user would actually say.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — "answer every question the PRD, prompt, or codebase can already settle", "resolve what's genuinely unknowable with timeboxed probes and manipulable prototypes", "gating progress on the user actually understanding", "Keeps its map and findings on disk" — with only minor coverage gaps, fitting the "several specific actions" anchor rather than the fully comprehensive 5.	4 / 5
Completeness	Explicitly answers both what (handle the parts of a plan that can't be decided by talking, probes, prototypes, disk map) and when ("Use this alongside… whenever the user answers 'I don't know yet'…", "Also use it when…") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases users would actually say — "I don't know yet", a plan "feels premature", "need to prototype first", "don't understand what the agent produced", "too big for one session", "previous plan turned out to be wrong", "resume a fogged effort" — covering multiple synonyms and phrasings.	5 / 5
Distinctiveness Conflict Risk	Carves a clear niche — "the third case: nobody knows yet" — explicitly contrasted against interview skills and mapping skills, with distinct triggers and minimal overlap risk.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 7 missing	Warning

	Total	15 / 16 Passed

Repository: MrToxy/claude-skills
Path: sightline/SKILL.md
Commit: 66c7102

Reviewed: 2 hours 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.