Tessl • The management layer for AI agent skills

SkillPublic

gemini-search

Search research papers via Gemini for broad literature discovery. Use when user says "gemini search", "gemini papers", "search with gemini", or wants AI-powered literature discovery beyond arXiv/Semantic Scholar indexes.

Quality

75%

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 ./skills/skills-codex/gemini-search/SKILL.md

Quality

Content

63%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 highly actionable with a complete MCP prompt, concrete fallback chain, and explicit output schema, and the workflow is well sequenced. Its main weaknesses are verbose time-sensitive model/version detail and the absence of any progressive disclosure — everything lives inline in one large file.

Suggestions

Move the model/version compatibility notes (gemini-cli v0.40+, auto-gemini-3 routing, downgrade behavior) into a short reference file or a clearly labeled 'legacy models' section, keeping only DEFAULT_MODEL and the override syntax in the body.

Make the CLI fallback fully executable by including the complete prompt in the `gemini -p` command instead of the '...same prompt as above...' placeholder.

Extract the full search prompt template and the sibling-skill comparison table into references/ files (e.g. PROMPT.md) and link to them, so SKILL.md stays a lean overview.

Dimension	Reasoning	Score
Conciseness	Mostly operational content, but the DEFAULT_MODEL constant carries a long time-sensitive parenthetical ('Required by mcp__gemini-cli__ask-gemini and gemini-cli v0.40+; explicit gemini-3-pro-preview can be silently downgraded... when capacity is exhausted') and sections like 'When to use which' re-explain things Claude can infer, so it sits at anchor 3 rather than the tighter anchor 4.	3 / 5
Actionability	The MCP call includes a complete copy-paste-ready prompt with an exact per-paper output format, plus concrete install, `codex mcp add`, and `gemini --version` commands; however the CLI fallback abbreviates the prompt ('...same prompt as above...') instead of giving the full executable command, which keeps it below anchor 5.	4 / 5
Workflow Clarity	Five clearly numbered steps with an explicit error-recovery chain (MCP first, CLI fallback, then alternative skills if both unreachable) and a setup verification command; but parsed results get no real validation step beyond 'note this but do not re-query', leaving minor checkpoint gaps versus anchor 5.	4 / 5
Progressive Disclosure	Section structure is clean, but there are no bundle files at all — the full search prompt template, the model/version compatibility notes, and the sibling-skill comparison table are all inlined in a ~200-line SKILL.md where they belong in one-level-deep reference files, matching anchor 3 rather than anchor 4.	3 / 5
	Total	14 / 20 Passed

Description

78%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 description with an explicit 'Use when' clause containing natural, verbatim trigger phrases and clear positioning against adjacent skills. Its only weakness is thin action coverage — it names one capability (broad literature search) where listing the concrete result artifacts would help.

Dimension	Reasoning	Score
Specificity	Names the domain and a single concrete action ('Search research papers via Gemini for broad literature discovery') but does not list several specific actions, so it falls at anchor 3 rather than 4.	3 / 5
Completeness	Explicitly answers both 'what' (search research papers via Gemini for broad literature discovery) and 'when' with concrete quoted trigger phrases, matching the anchor-5 example structure exactly; nothing is missing for a higher score and there is no higher anchor.	5 / 5
Trigger Term Quality	Includes three natural verbatim trigger phrases ('gemini search', 'gemini papers', 'search with gemini') plus a situational trigger ('beyond arXiv/Semantic Scholar indexes'), but misses common variations users would say such as 'find papers' or 'literature search', placing it just below the comprehensive coverage of anchor 5.	4 / 5
Distinctiveness Conflict Risk	Explicitly positions against sibling skills ('beyond arXiv/Semantic Scholar indexes') and uses distinct Gemini-named triggers, but generic 'search research papers' requests could still overlap with arXiv/Semantic Scholar skills, leaving minor overlap risk rather than the minimal risk of anchor 5.	4 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/skills-codex/gemini-search/SKILL.md
Revision: 18bceb3

Reviewed: 29 days ago

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.

gemini-search

Table of Contents

Is this your skill?