CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gemini-cli-bridge

Integrate with Google's Gemini CLI for delegating tasks to Gemini. Use when you need an alternative AI perspective, Google Search grounding, parallel task execution, or want to leverage Gemini's 1M token context window. Enables "Claude calling Gemini" patterns for diverse AI collaboration.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./planning/done/gemini-cli/SKILL.md

The canonical home for this skill is gemini-cli-bridge in Raudbjorn/TTTRPS

SKILL.md
Quality
Evals
Security

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.

A well-structured, mostly concrete skill body with real commands, config, and error-recovery guidance. Its weaknesses are redundant restatement of Gemini's strengths across four sections and pseudo-syntax workflow examples that aren't directly executable.

Suggestions

Consolidate the overlapping Gemini-strengths content: merge 'Key Differences' and 'Best Practices' points into the 'When to Use' section so search/big-context benefits are stated once.

Replace the pseudo-blocks in 'Example Workflows' ('[gemini_prompt] prompt: ...') with actual executable invocations, e.g. the real gemini_prompt tool call or a concrete gemini CLI command per workflow.

Move the Authentication methods and MCP/CLI configuration detail into a references/ file, keeping SKILL.md as a lean overview with one-level-deep links.

DimensionReasoningScore

Conciseness

The 'When to Use' list, 'Key Differences' table, 'Best Practices', and 'Example Workflows' each independently restate the same search/big-context strengths ('Gemini excels at queries needing real-time info', 'Leverage Search'), so there is more than minor redundancy. It stays at 3 rather than 2 because most sections carry concrete, non-obvious facts (rate limits, auth methods, commands) rather than concept explanations Claude already knows.

3 / 5

Actionability

Mostly executable guidance: concrete bash invocations ('gemini -p "..." --output-format json'), pipe examples, a JSON MCP config block, env-var exports, and an install command. It falls short of 5 because the Example Workflows use non-executable pseudo-blocks ('[gemini_search] query: ...') rather than real tool-call forms, and MCP tool parameters like working_dir/yolo_mode are never shown in an actual invocation.

4 / 5

Workflow Clarity

Usage patterns are clearly ordered (MCP recommended, then direct CLI, then library) and an Error Handling section provides explicit recovery steps ('Not Found: Install with npm install -g @google/gemini-cli', 'Auth Failed: Run gemini interactively'). Minor gaps: no explicit guidance for verifying which method is available before use, so it does not reach 5; the error-recovery steps keep it above 3.

4 / 5

Progressive Disclosure

A single-file skill (~165 lines) with no bundle files, well-organized into scannable sections (Usage Patterns, Configuration, Error Handling, Authentication). Most content is appropriately placed at overview level, though the detailed auth methods and example workflows could move to a reference file, and the under-50-line exception for a 5 does not apply at this length.

4 / 5

Total

15

/

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 what+when structure and concrete trigger phrases. The main defect is second-person voice ('Use when you need...'), which the rubric penalizes, plus slightly broad buzz-phrasing around AI collaboration that weakens distinctiveness.

Suggestions

Rewrite the trigger clause in third person to remove the specificity penalty, e.g. 'Use when an alternative AI perspective, Google Search grounding, parallel task execution, or a 1M-token context window is needed.'

Replace 'diverse AI collaboration' with a concrete phrase such as 'delegate tasks to Gemini' to tighten distinctiveness against generic multi-AI skills.

DimensionReasoningScore

Specificity

The description lists several concrete actions ('delegating tasks to Gemini', 'Google Search grounding', 'parallel task execution', '1M token context window'), which matches the anchor-4 level, but it uses second-person voice ('Use when you need', 'or want to leverage'), so the rubric's voice penalty reduces it by one. It is not level 2 because the actions go well beyond minimal or generic.

3 / 5

Completeness

Explicitly answers both: what ('Integrate with Google's Gemini CLI for delegating tasks to Gemini') and when ('Use when you need an alternative AI perspective, Google Search grounding, parallel task execution, or want to leverage Gemini's 1M token context window'). The 'when' clause includes concrete trigger phrases, matching the top anchor; it does not fall to 4 since the when-guidance is already explicit and specific.

5 / 5

Trigger Term Quality

Good natural keyword coverage — 'Gemini', 'Gemini CLI', 'Google Search', 'parallel task execution', '1M token context window' — terms a user would plausibly say when needing this skill. It falls short of 5 because common variations like 'ask Gemini', 'delegate to Gemini', or 'second opinion' are absent and 'diverse AI collaboration' is buzzword-adjacent.

4 / 5

Distinctiveness Conflict Risk

The Gemini CLI delegation niche is mostly distinct with clear triggers, but phrases like 'alternative AI perspective' and 'diverse AI collaboration' create minor overlap risk with generic multi-AI or web-search delegation skills. Not 5 because of this residual broad phrasing; not 3 because the core trigger (Gemini) is highly specific.

4 / 5

Total

16

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

Raudbjorn/TTTTRPS

Path

planning/done/gemini-cli/SKILL.md

Reviewed

Table of Contents

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.