research

Research phase of RPI methodology. Spawns parallel subagents for codebase exploration AND web/pattern research, then synthesizes findings for user review. Produces compact research artifact. Use at the start of non-trivial features.

Quality

69%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./tiles/crafter/skills/research/SKILL.md

Quality

Content

67%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 a well-structured, mostly actionable workflow with clear sequencing, a review feedback loop, and properly signaled one-level-deep references to real bundle files. The main weakness is conciseness: several sections repeat the same STOP/next-steps content, and the rationale sections could be tightened.

Suggestions

Collapse the duplicate STOP guidance: keep the exact output block in step 6 and remove the redundant 'STOP — Phase Complete' and 'After Research' restatements.

Tighten or remove the 'Context Compaction' section — the /clear rationale is already covered in the required output block.

Make the dispatch example closer to literal tool-call syntax (or label it explicitly as illustrative) to remove ambiguity about how to invoke the Task tool.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with useful tables and a clear workflow, but contains redundancy: the STOP directive appears both in step 6 and again as a full 'STOP — Phase Complete' section, and 'After Research' restates steps 4-6, plus a 'Context Compaction' rationale that could be trimmed.	3 / 5
Actionability	Provides concrete tool params (run_in_background: true, TaskOutput block: true/false), exact output paths, and a copy-paste-ready final markdown block; the dispatch example is illustrative rather than literal tool-call syntax, a minor gap.	4 / 5
Workflow Clarity	Six steps are clearly sequenced with a user-review feedback loop ('If the user requests edits -> update... Repeat until the user approves') and gap handling in step 3; only minor validation gaps remain (no explicit 'all agents complete before synthesis' gate beyond polling).	4 / 5
Progressive Disclosure	Body is an overview with well-signaled one-level-deep references to real bundle files ([agent prompts](references/agent-prompts.md) and [research artifact template](references/template.md)); content is appropriately split, with minor organization gaps.	4 / 5
	Total	15 / 20 Passed

Description

71%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 specific and covers both what the skill does and when to use it, with concrete actions and natural trigger terms. Its main weakness is distinctiveness: the trigger vocabulary is broad and risks overlapping with general exploration or built-in agents.

Suggestions

Narrow the triggers to reduce conflict with the Explore agent, e.g. 'research before planning a feature' rather than generic 'investigate'.

Add a concrete trigger phrase to the 'when' clause, e.g. 'Use when starting a non-trivial feature and the user mentions RPI, planning, or pre-implementation research.'

Include a synonym cluster (e.g. 'look into', 'understand how X works') to broaden natural-term coverage.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Spawns parallel subagents for codebase exploration AND web/pattern research', 'synthesizes findings', 'Produces compact research artifact') with only minor gaps in coverage.	4 / 5
Completeness	Explicitly answers both what (research methodology producing a compact artifact) and when ('Use at the start of non-trivial features'); the 'when' is present and explicit but could be more specific with concrete trigger phrases.	4 / 5
Trigger Term Quality	Includes natural terms users would say ('research', 'explore codebase', 'investigate', 'non-trivial features') with good coverage, though a few common synonyms ('look into', 'understand how') are missing.	4 / 5
Distinctiveness Conflict Risk	Has a niche ('Research phase of RPI methodology') but broad triggers like 'research' and 'explore codebase' overlap with general exploration skills; the skill itself even admits overlap with the built-in Explore agent.	3 / 5
	Total	15 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: ericjohnolson/tessl-test
Path: tiles/crafter/skills/research/SKILL.md
Commit: 2a98cc1

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.