functional-area-resolver

Compress an agent's routing file (RESOLVER.md or AGENTS.md) by converting granular skill-per-row tables into functional-area dispatchers. Each area lists sub-skills in a "(dispatcher for: ...)" clause. The LLM reads one area entry and routes to the correct sub-skill. Proven via held-out A/B eval: dispatcher pattern outperforms naive pipe-table compression.

Quality

72%

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 ./skills/functional-area-resolver/SKILL.md

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with a well-sequenced, validated workflow, but it is a long monolith that mixes operational how-to with eval methodology, citations, and version-specific time markers that inflate tokens and would be better split into referenced files.

Suggestions

Move the A/B Eval Results, Methodology caveats, and Prior work/citations sections into a separate EVAL.md (or a references/ file) and leave a one-line pointer in SKILL.md; this trims the inlined eval content driving the conciseness and progressive-disclosure scores down.

Consolidate the scattered time-sensitive version and date markers (v0.32.3.0, v0.31.7, v0.33.x, v0.36.x, 2026-05-11) into a single 'Compatibility / versions' or 'Deprecated' section so they stop penalizing inline conciseness.

Trim the four-point 'What the data shows' narrative to the single load-bearing claim (the dispatcher clause is the routing signal); the tables already convey the rest.

Dimension	Reasoning	Score
Conciseness	The core how-to (Steps 1-7, Anti-Patterns, Contract) is efficient, but the body pads ~318 lines with full A/B eval tables, four-point data analysis, arXiv citations, and inline time-sensitive version/date markers (v0.32.3.0, 2026-05-11, v0.33.x) not isolated in a deprecated section — matching the anchor that is mostly efficient but could be tightened.	2 / 3
Actionability	Guidance is fully executable: exact commands ('gbrain routing-eval --json', the complete 'node harness.mjs --variants-dir ... --model opus --parallel 3 --yes'), concrete thresholds (95%, 12KB), an entry template, and copy-paste bash blocks, matching the copy-paste-ready anchor.	3 / 3
Workflow Clarity	Steps 1-7 are clearly sequenced with explicit validation checkpoints — Gate 1 (structural, 95%, revert-and-tune) and Gate 2 (LLM A/B, 95%, revert-and-tune) — providing the feedback loops the rubric rewards for destructive/batch edits.	3 / 3
Progressive Disclosure	No bundle files exist and the ~318-line SKILL.md is monolithic: eval tables, methodology caveats, and prior-work citations are inlined where a separate EVAL.md/REFERENCES.md would fit, while the repo-path references are signaled but point outside the skill rather than to bundle files.	2 / 3
	Total	10 / 12 Passed

Description

67%

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 and well-differentiated but reads more like a methodology summary than a trigger-oriented skill description: it explains the mechanism and eval provenance but omits an explicit 'when to use' clause and natural user phrasings.

Suggestions

Add an explicit 'Use when...' clause to the description stating when this skill should activate (e.g. 'Use when an agent's routing file (RESOLVER.md/AGENTS.md) exceeds ~12KB and skill-per-row tables are consuming context budget').

Fold 1-2 natural trigger phrases the user would actually say ('resolver too big', 'compress agents.md') into the description text itself, not just the triggers field, to improve trigger-term coverage.

Trim the eval-provenance sentence ('Proven via held-out A/B eval...') from the description; that evidence belongs in the body, and it does not aid routing.

Dimension	Reasoning	Score
Specificity	The description names multiple concrete actions — 'Compress an agent's routing file (RESOLVER.md or AGENTS.md)', 'converting granular skill-per-row tables into functional-area dispatchers', and 'lists sub-skills in a (dispatcher for: ...) clause' — matching the anchor that lists multiple specific concrete actions.	3 / 3
Completeness	It clearly answers 'what' (compress routing files into functional-area dispatchers) but has no explicit 'Use when...' trigger clause; the 'when' is only implied, so per the judging guideline a missing explicit trigger guidance caps completeness at 2.	2 / 3
Trigger Term Quality	It includes relevant technical terms ('routing file', 'RESOLVER.md', 'AGENTS.md', 'functional-area dispatchers') but omits the common natural phrasings a user would say (e.g. 'resolver too big', 'shrink routing table'); those live only in the triggers field, which the description rubric does not score.	2 / 3
Distinctiveness Conflict Risk	The niche is narrow and clearly distinct — compressing routing tables into functional-area dispatchers — with specific signals (RESOLVER.md/AGENTS.md, dispatcher clause) unlikely to trigger for unrelated skills.	3 / 3
	Total	10 / 12 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: garrytan/gbrain
Path: skills/functional-area-resolver/SKILL.md
Commit: 3fafb69

Reviewed: about 20 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.