fractal

Delegate analysis- and context-heavy work to Fractal, an agentic CLI powered by a self-harnessed Recursive Language Model (predict-rlm), by running it non-interactively (fractal -p). Reach for it when a task needs reasoning over a large or deep codebase, synthesizing an answer across many files, auditing, or open-ended investigation — work that would otherwise flood your own context. The RLM reasons over context programmatically (no context rot) and returns a distilled answer. Use when asked to run Fractal headless, script it, call it from CI or another agent, or offload a heavy analysis/large-context task. Less suited to trivial single-file edits you can do directly.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

78%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 highly actionable with executable commands, a precise output contract, and well-structured sections, but it is undermined by a dead reference to a RECIPES.md file that is not present in the bundle. Conciseness and workflow sequencing are solid with minor room to tighten.

Suggestions

Add the missing RECIPES.md file (referenced in the closing line and in the --json flag entry as RECIPES.md#structured-output), or remove those references — broken links promise navigation that does not exist.

Collapse the install, preflight, and run steps into a single numbered checklist with an explicit validate-then-run feedback loop (e.g. 'if preflight fails, run fractal config setup / fix Docker, then re-check') to make the workflow sequence and recovery path explicit.

Trim the 'When to reach for it' section, which largely restates the frontmatter description, to reduce redundancy and reclaim token budget.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and assumes competence, but the 'When to reach for it' section overlaps heavily with the frontmatter description and the RLM-mechanism explanation is slightly more than needed, so a few lines could be trimmed.	4 / 5
Actionability	Fully executable, copy-paste-ready guidance throughout: an install one-liner, `fractal -p "..." --workspace /path`, preflight checks, a complete output-contract table with exit codes, and concrete flag syntax like `out=$(fractal -p "..." 2>err.log)`.	5 / 5
Workflow Clarity	A clear install → preflight → run → check-exit-code sequence exists with a validation checkpoint (`fractal config status`, `docker info`), but the steps are distributed across sections rather than presented as one explicit checklist, and there is no fix-and-retry feedback loop for failed runs.	4 / 5
Progressive Disclosure	Sections are well organized and the body signals a one-level-deep reference to RECIPES.md for advanced recipes, but that referenced file (cited twice, including `RECIPES.md#structured-output`) does not exist in the bundle, so the promised navigation is broken.	3 / 5
	Total	16 / 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.

The description is strong: it clearly states what Fractal does and when to reach for it, with concrete trigger language and a distinct tool niche. Third-person voice is maintained throughout and there is no vague fluff.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — 'running it non-interactively (fractal -p)', 'reasoning over a large or deep codebase, synthesizing an answer across many files, auditing, or open-ended investigation' — but they read more as use-case triggers than distinct functional capabilities, leaving minor coverage gaps versus the comprehensive anchor.	4 / 5
Completeness	Explicitly answers both what ('Delegate analysis- and context-heavy work to Fractal ... by running it non-interactively') and when ('Use when asked to run Fractal headless, script it, call it from CI or another agent, or offload a heavy analysis/large-context task') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural trigger phrases — 'run Fractal headless', 'script it', 'call it from CI or another agent', 'offload a heavy analysis/large-context task' — with good coverage; a few common synonyms (e.g. 'review', 'trace') are only implied rather than named.	4 / 5
Distinctiveness Conflict Risk	Names a specific tool ('Fractal, an agentic CLI powered by a self-harnessed Recursive Language Model (predict-rlm)') with distinct triggers, creating a clear niche and minimal conflict risk with other skills.	5 / 5
	Total	18 / 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: 2 missing	Warning

	Total	15 / 16 Passed

Repository: jasonkneen/tiny-world-builder
Path: .agents/skills/fractal/SKILL.md
Commit: 6a9a551

Reviewed: 4 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.