Claude Code
- Surfaces represented
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context represented
- Not established
Practical agent comparison
Claude Code vs Gemini CLI
Both products document hierarchical project guidance, coding tools, filesystem changes, shell execution, and permission controls. Claude Code uses scoped CLAUDE.md files and repository auto memory. Gemini CLI uses hierarchical GEMINI.md context and documents planning and web tools; the reviewed evidence does not establish comparable persistent memory. Their confirmation, sandbox, and context-assembly mechanisms differ, so preserve the destination product's control model when migrating workflows.
- Reviewed
- Products
- Claude Code · Gemini CLI
- Decision criteria
- 9 reviewed questions
Comparison orientation
Four reviewed dimensions at a glance
Each dimension links to its detailed evidence row. State labels describe the selected evidence.
Left productClaude CodeOpen product record
- Documented 01Repository / project instructionsOpen detailed evidence Documented
- Documented 02Memory / persistent contextOpen detailed evidence Not established
- Documented 03Planning / work decompositionOpen detailed evidence Documented
- Documented 04Tools / tool callingOpen detailed evidence Documented
Who this comparison is for
How do Claude Code and Gemini CLI assemble documented instructions and controls?
Developers and engineering teams comparing multi-surface Claude Code workflows with terminal-based Gemini CLI automation.
Evidence-bounded scope
Product, surface, and model context
These summaries combine only the scope stated in the reviewed answers. Each criterion keeps its narrower limits.
Gemini CLI
- Surfaces represented
- Gemini CLI
- Model context represented
- Not established
Practical behavior comparison
What the selected evidence establishes
“Not established” records an evidence gap. It does not mean the product lacks the capability.
| Criterion | Claude Code | Gemini CLI | Interpretation |
|---|---|---|---|
| Repository / project instructions Repository guidance must be placed in the target agent's documented instruction hierarchy. | Documented Scoped CLAUDE.md hierarchy
| Documented Hierarchical GEMINI.md context
| Read practical implication |
| Memory / persistent context Durable workflow context should not be assumed to persist across agent environments. | Documented Repository-scoped auto memory
| Not established Persistent memory semantics not established
| Read practical implication |
| Planning / work decomposition Planning assumptions shape how teams structure, review, and resume multi-step work. | Documented Plan mode before source edits
| Documented Planning appears in documented tools
| Read practical implication |
| Tools / tool calling A portable workflow must use only the tools established by the selected evidence. | Documented Read, edit, execute, and external tools
| Documented Filesystem, execution, planning, MCP, and web
| Read practical implication |
| Filesystem / file modification Repository reads and mutations require the correct workspace and review boundary. | Documented Reads and edits codebase files
| Documented Filesystem tools with confirmation for mutation
| Read practical implication |
| Shell / terminal execution Command-heavy automation depends on the target harness's execution controls. | Documented Runs commands through codebase tools
| Documented Shell tools with confirmation and isolation
| Read practical implication |
| Browsing / network access Current external information cannot be assumed available when browsing scope is unestablished. | Documented WebSearch and WebFetch
| Documented Web tools in official tool reference
| Read practical implication |
| Approvals / permissions Consent, confirmation, and isolation rules determine which delegated actions can proceed. | Documented Deny, ask, allow, and permission modes
| Documented Confirmation, sandbox, and trusted folders
| Read practical implication |
| Context scope / inheritance Instructions and tools remain dependable only when their surface, model, and configuration boundaries are preserved. | Documented Managed, user, project, local, and memory scopes
| Documented Global, workspace, just-in-time, and dynamic context
| Read practical implication |
Criterion-by-criterion interpretation
Scope, limitations, and evidence
Behavior Intelligence supplies the factual answers. Editorial implications remain separate.
01
Open evidence across six products How do AI agents use repository or project instructions?
Why it matters. Repository guidance must be placed in the target agent's documented instruction hierarchy.
Claude Code
DocumentedClaude Code documents managed, user, project, and local CLAUDE.md instruction scopes for coding-agent work.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- Not established
Limits and uncertainty
- The complete default base system prompt is not established.
Gemini CLI
DocumentedGemini CLI documents global, workspace, and just-in-time GEMINI.md files as hierarchical project and task context.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Not established
Limits and uncertainty
- The context hierarchy is distinct from the replaceable built-in system prompt.
02
Open evidence across six products Does the product document memory or persistent context?
Why it matters. Durable workflow context should not be assumed to persist across agent environments.
Claude Code
DocumentedClaude Code documents repository-scoped auto memory that stores selected learnings separately from human-authored CLAUDE.md instructions.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- Not established
Limits and uncertainty
- Auto memory is context, not hard policy enforcement.
Gemini CLI
Not establishedThe selected Gemini CLI tools reference names memory among its categories, but does not establish enough persistence or scope behavior for this question.
- Surface
- Not established
- Model context
- Not established
- Configuration / version
- Not established
Limits and uncertainty
- A tool category alone is not treated as proof of cross-session persistence.
No source link is attached because the reviewed evidence does not establish this answer.
03
Open evidence across six products How is planning or work decomposition documented?
Why it matters. Planning assumptions shape how teams structure, review, and resume multi-step work.
Claude Code
DocumentedPlan mode researches and proposes changes before source edits. It can run exploratory shell commands and write a plan. Outside sessions with bypass permissions available, source edits remain blocked until plan approval. Command review depends on mode, settings, and version; approval exits planning into the selected execution mode. CLI evidence is not a universal read-only sandbox claim.
- Surface
- Claude Code terminal CLI
- Model context
- Not established
- Configuration / version
- Current product documentation consulted 2026-09-14; enabled mode, settings, and deployment still determine availability.
Limits and uncertainty
- This claim does not update the dated prompt snapshot or establish every runtime configuration.
Gemini CLI
DocumentedThe selected Gemini CLI tools reference includes planning among the documented tool categories available to the CLI.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Not established
Limits and uncertainty
- The selected evidence does not establish one fixed planning sequence across configurations.
04
Open evidence across six products What tools or tool-calling capabilities are documented?
Why it matters. A portable workflow must use only the tools established by the selected evidence.
Claude Code
DocumentedClaude Code documents tools for reading code, editing files, running commands, and connecting to external tools.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- Permission rules and modes can constrain tool execution.
Limits and uncertainty
- The selected evidence does not establish that every deployment exposes every external tool.
Gemini CLI
DocumentedGemini CLI's selected official reference covers filesystem, execution, planning, memory, MCP, and web tool categories.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Enabled tools and security posture can vary with configuration.
Limits and uncertainty
- The selected evidence does not establish one immutable tool set.
05
Open evidence across six products Can the agent read or modify files?
Why it matters. Repository reads and mutations require the correct workspace and review boundary.
Claude Code
DocumentedClaude Code's official overview says the product reads code, edits files, and runs commands.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- Permission rules distinguish read-only and mutating actions.
Limits and uncertainty
- Access remains bounded by the client environment and configured permissions.
Gemini CLI
DocumentedGemini CLI documents filesystem tools, with mutating file operations subject to confirmation and configured security boundaries.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Sandbox and trusted-folder settings can change the execution boundary.
Limits and uncertainty
- Tool availability varies by configuration and version.
06
Open evidence across six products Can the agent execute shell or terminal commands?
Why it matters. Command-heavy automation depends on the target harness's execution controls.
Claude Code
DocumentedClaude Code's official overview documents command execution as part of codebase operations.
- Surface
- Claude Code terminal CLI
- Model context
- Not established
- Configuration / version
- Permission rules and modes determine when command execution is allowed or asks for approval.
Limits and uncertainty
- The exact command environment is deployment-specific.
Gemini CLI
DocumentedGemini CLI documents shell execution tools whose requests pass through confirmation and configured security policies.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Sandboxing and trusted-folder settings can constrain execution.
Limits and uncertainty
- The enabled tool set is configuration-dependent.
07
Open evidence across six products Can the product browse or access web and network resources?
Why it matters. Current external information cannot be assumed available when browsing scope is unestablished.
Claude Code
DocumentedWebSearch returns search titles and URLs; WebFetch retrieves and processes a page, with lossy summarization and possible truncation. Tool/domain rules and provider support apply. Amazon Bedrock does not expose the server-side search tool; an Azure-hosted Foundry deployment does not support it. Web tools do not establish arbitrary network access.
- Surface
- Claude Code terminal CLI
- Model context
- Not established
- Configuration / version
- Current product documentation consulted 2026-09-14; enabled mode, settings, and deployment still determine availability.
Limits and uncertainty
- This claim does not update the dated prompt snapshot or establish every runtime configuration.
Gemini CLI
DocumentedGemini CLI's selected official tools reference includes web capabilities among the available tool categories.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Availability depends on enabled tools and security configuration.
Limits and uncertainty
- The selected evidence does not establish unrestricted network access.
08
Open evidence across six products What approval or permission controls are documented?
Why it matters. Consent, confirmation, and isolation rules determine which delegated actions can proceed.
Claude Code
DocumentedClaude Code documents deny, ask, and allow rules with precedence, plus permission modes that constrain tool execution.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- Rules can be managed at different settings scopes and by the client environment.
Limits and uncertainty
- The selected evidence does not establish every deployment policy.
Gemini CLI
DocumentedGemini CLI documents confirmation for mutating file and shell tools, with sandboxing and trusted-folder boundaries.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- Security policy and isolation settings shape the actual boundary.
Limits and uncertainty
- The selected evidence does not establish every enterprise policy.
09
Open evidence across six products How is context scoped or inherited across products and agents?
Why it matters. Instructions and tools remain dependable only when their surface, model, and configuration boundaries are preserved.
Claude Code
DocumentedClaude Code documents managed, user, project, and local CLAUDE.md scopes plus separate repository-scoped auto memory.
- Surface
- Claude Code terminal CLI · Claude Code IDE integrations
- Model context
- Not established
- Configuration / version
- The reviewed evidence covers terminal and IDE-oriented product surfaces with a selectable Claude model.
Limits and uncertainty
- The complete default system prompt is not established.
Gemini CLI
DocumentedGemini CLI documents global, workspace, and just-in-time GEMINI.md context plus dynamic injection of enabled skills, subagents, and tools.
- Surface
- Gemini CLI
- Model context
- Not established
- Configuration / version
- The product uses a configurable Gemini model and can fully replace its built-in system prompt.
Limits and uncertainty
- No single immutable runtime assembly is claimed.
Workflow guidance
Migration considerations
These implications follow from the reviewed criteria. They do not express a product preference.
Example: move repository guidance without copying learned state
Separate the policy to preserve from context that needs a fresh review before a Claude Code workflow moves to Gemini CLI.
- Suppose project CLAUDE.md says to validate generated API clients, user guidance prefers concise responses, and Claude Code auto memory contains a note about an old generation command. Treat these as three different inputs.
- Put the team validation requirement in the workspace GEMINI.md context and the personal preference in its documented global context. Use the project’s actual generation command; do not promote an auto-memory note to team policy without checking it.
- Inspect Gemini CLI just-in-time context and enabled tools, skills, and subagents as well as GEMINI.md. Its dynamic prompt inputs are not reproduced by copying CLAUDE.md. A configured full system-prompt replacement also needs separate inspection.
- Require a proposed file list and validation plan before approving edits. Both products document planning, but the selected Gemini tool evidence does not prove Claude Code’s Plan-mode approval flow applies there. Set Gemini confirmation and isolation controls independently.
- Handoff condition: the destination identifies the intended generation command and team rule, and the operator verifies its execution boundary. Keep the old memory note as an unresolved handoff question until checked; no equivalent Gemini automatic-memory behavior is established here.
Make planning and persistence requirements explicit
Both products establish planning, but the selected Gemini CLI record does not establish an equivalent to Claude Code auto memory. Carry task state explicitly and review the destination execution controls before work resumes.
Carry the destination control model
Filesystem and shell workflows must honor Claude Code permission rules or Gemini CLI confirmation and isolation rather than a generic approval assumption.
Recheck network-dependent steps
A web-dependent step can fit either product, but the target must expose the required tools. On Claude Code, verify provider support for WebSearch and domain permission for WebFetch; on Gemini CLI inspect enabled web tools and security settings.
Comparison-specific evidence
Differences in available evidence
Claude Code and Gemini CLI combine model context, project instructions, tools, and harness controls in different ways. Their official documentation differs most clearly in base-prompt handling, dynamic assembly, and project-context scope. The analysis below adds prompt and product context to the shared behavior criteria.
- 01
Gemini CLI documents prompt export and replacement
Gemini CLI exposes workflows to export its current built-in prompt and completely replace it. No equivalent full-prompt workflow is established for Claude Code in the selected sources.
- 02
Both document hierarchical project context
Claude Code uses scoped CLAUDE.md files; Gemini CLI uses global, workspace, and just-in-time GEMINI.md context.
- 03
Dynamic context works differently
Claude Code documents repository auto memory. Gemini CLI documents dynamic substitution of skills, subagents, and enabled tools into custom prompts.
Source-specific detail
Product and prompt evidence
These criteria cover prompt visibility, product identity, and other evidence specific to this comparison.
| Criterion | Claude Code | Gemini CLI |
|---|---|---|
| Product and model roleEditorial interpretation. Both are coding-agent harnesses with a selectable or configurable model rather than one product-equals-model identity. | documented Claude Code spans terminal, IDE, desktop, and web surfaces with a separately selected Claude model. | documented Gemini CLI is an open-source terminal agent whose configured Gemini model is separate from its instruction assembly. |
| Base-prompt handlingEditorial interpretation. Official Gemini CLI documentation exposes a prompt workflow that the selected Claude Code documentation does not establish. | not established The selected official sources do not publish or expose a complete default Claude Code system prompt. | documented GEMINI_WRITE_SYSTEM_MD can export the current built-in prompt, while GEMINI_SYSTEM_MD can replace it completely. |
| Project instructionsEditorial interpretation. Both products document hierarchical instruction files, but the filenames and scope models differ. | documented CLAUDE.md instructions can be managed, user-scoped, project-scoped, or local. | documented GEMINI.md context can load globally, from a workspace hierarchy, and just in time from relevant directories. |
| Dynamic instruction assemblyEditorial interpretation. Gemini CLI's selected docs are more explicit about named dynamic substitutions; Claude Code's selected docs focus on scoped files and memory. | documented Claude Code assembles scoped CLAUDE.md context and repository memory, but this source set does not establish an equivalent named substitution template. | documented Custom prompts can substitute available skills, subagents, and enabled tool names, making the resolved prompt configuration-dependent. |
| Tools and permission boundariesEditorial interpretation. Both harnesses mediate agent actions outside model text, with different documented rule vocabularies. | documented Claude Code documents code, file, command, and external-tool operations plus deny, ask, and allow permission rules. | documented Gemini CLI documents filesystem, execution, planning, memory, MCP, and web tools with confirmation and sandbox boundaries. |
| Persistent memoryEditorial interpretation. Claude Code documents a repository-scoped auto-memory store. The selected Gemini CLI source names a memory tool but does not establish an equivalent automatic store. | documented Auto memory stores selected learnings separately from human-authored CLAUDE.md instructions. | not established The selected Gemini CLI tools reference includes memory capability, but does not establish an equivalent repository auto-memory mechanism. |
Using the evidence
How to read this evidence
Choose the evidence boundary before comparing prompts
Gemini CLI supports explicit export and replacement claims; Claude Code's selected sources support architecture analysis without a complete default-prompt claim.
Map repository guidance by scope
Both products support persistent project context, but CLAUDE.md and GEMINI.md have different documented hierarchies and should not be treated as interchangeable names.
Separate prompt assembly from enforcement
Dynamic substitutions and instruction files shape context; tool confirmations, permission rules, and sandboxing shape what actions proceed.
Primary-source trail
Sources used in this comparison
Each source links to its product evidence, with provenance and publication limits preserved.
Claude Code
- Tier A · vendor-publishedClaude Code permission modes — Plan mode
- Tier A · vendor-publishedClaude Code tools reference — WebSearch and WebFetch
- Tier A · vendor-publishedClaude Code overview
- Tier A · vendor-publishedHow Claude remembers your project
- Tier A · vendor-publishedClaude Code permissions
Gemini CLI
- Tier A · vendor-publishedGemini CLI system-prompt override documentation
- Tier A · vendor-publishedGemini CLI GEMINI.md context documentation
- Tier A · vendor-publishedGemini CLI tools reference
Deeper cross-agent context
Behavior questions in this comparison
Open the six-product answer for broader evidence and scope.
Repository / project instructionsHow do AI agents use repository or project instructions?Memory / persistent contextDoes the product document memory or persistent context?Planning / work decompositionHow is planning or work decomposition documented?Tools / tool callingWhat tools or tool-calling capabilities are documented?Filesystem / file modificationCan the agent read or modify files?Shell / terminal executionCan the agent execute shell or terminal commands?Browsing / network accessCan the product browse or access web and network resources?Approvals / permissionsWhat approval or permission controls are documented?Context scope / inheritanceHow is context scoped or inherited across products and agents?
Continue comparing