Curated CLAUDE.md system prompt variations for different coding styles and strictness levels — conservative, balanced, and aggressive modes with detailed trade-off notes.
[!tip]
NEW (June 12, 2026): We've greatly expanded this list with many more of Claude Code's prompts—from 350 to 515 (+165)—our most complete coverage yet.
This repository contains an up-to-date list of all Claude Code's various system prompts and their associated token counts as of Claude Code v2.1.296 (October 9th, 2026). It also contains a CHANGELOG.md for the system prompts across 312 versions since v2.0.14. From the team behind Piebald.
This repository is updated within minutes of each Claude Code release. See the changelog, and follow @PiebaldAI on X for a summary of the system prompt changes in each release.
[!note]
⭐ Star this repository to get notified about new Claude Code versions. For each new Claude Code version, we create a release on GitHub, which will notify all users who've starred the repository.
Why multiple "system prompts?"
Claude Code doesn't just have one single string for its system prompt.
Instead, there are:
Large portions conditionally added depending on the environment and various configs.
Descriptions for builtin tools like Write, Bash, and TodoWrite, and some are fairly large.
Separate system prompts for builtin agents like Explore and Plan.
Numerous AI-powered utility functions, such as conversation compaction, CLAUDE.md generation, session title generation, etc. featuring their own systems prompts.
The result—500+ strings that are constantly changing and moving within a very large minified JS file.
[!important]
Want to modify a particular piece of the system prompt in your own Claude Code installation? Use tweakcc. It—
lets you customize the individual pieces of the system prompt as markdown files, and then
patches your npm-based or native (binary) Claude Code installation with them, and also
provides diffing and conflict management for when both you and Anthropic have conflicting modifications to the same prompt file.
Extraction
This repository contains the system prompts extracted using a script from the latest npm version of Claude Code. As they're extracted directly from Claude Code's compiled source code, they're guaranteed to be exactly what Claude Code uses. If you use tweakcc to customize the system prompts, it works in a similar way—it patches the exact same strings in your local installation as are extracted into this repository.
Prompts
Note that some prompts contain interpolated bits such as builtin tool name references, lists of available sub agents, and various other context-specific variables, so the actual counts in a particular Claude Code session will differ slightly—likely not beyond ±20 tokens, however.
Agent Prompt: /code-review minimal mode (632 tks) - Minimal /code-review prompt that performs one careful diff pass and reports up to fifteen concrete correctness findings.
Agent Prompt: /schedule slash command (4469 tks) - Guides the user through scheduling, updating, listing, or running remote Claude Code agents on cron triggers via the Anthropic cloud API.
Agent Prompt: /simplify slash command (362 tks) - Instructions for the /simplify slash command that reviews changed code for reuse, simplification, efficiency, and altitude cleanups, then applies the fixes.
Agent Prompt: Artifact type creation slash command (347 tks) - Directs an Artifact creation slash command to discover the named published Artifact type, create from its type URL, and follow the returned type instructions.
Utilities
Agent Prompt: Agent Hook (289 tks) - Evaluates agent hook conditions against the codebase and, when available, the conversation transcript, then returns a structured pass/fail result.
Agent Prompt: Artifact comment thread analyst (750 tks) - Instructs a read-only agent to analyze one Artifact comment thread and return a constrained analysis brief for a separate composer.
Agent Prompt: Artifact comment thread background directive (300 tks) - Directs a forked background agent to handle one Artifact comment thread on its own, leaving the main session's pending work alone and treating thread and artifact content as material rather than instructions.
Agent Prompt: Artifact editor thread follow-up (328 tks) - Delivers a tagged Artifact thread message to the active editor worker, requiring page-scoped edits and republishing for relevant requests and no changes for unrelated ones.
Agent Prompt: Auto mode rule reviewer (292 tks) - Reviews and critiques user-defined auto mode classifier rules for clarity, completeness, conflicts, and actionability.
Agent Prompt: Away summary generation (73 tks) - Prompts a no-tools away-summary generation run to recap the goal, current task, and next action when the user returns.
Agent Prompt: Background agent state classifier chat-thread relay rule (1006 tks) - Classifier addendum for sessions that reach people only through posts in a chat thread: a named person counts as the user, an open ask stays blocked even beside a scheduled check-in, and third-party waits stay done.
Agent Prompt: Background job agent instructions (581 tks) - Instructs the built-in background job agent to narrate progress, restate tool results, and emit explicit result, needs input, or failed status signals.
Agent Prompt: Claude Code guide (408 tks) - Describes when to use the Claude Code guide agent for Claude Code, Agent SDK, Claude API, and Claude Tag questions.
Agent Prompt: Claude Test author (346 tks) - Subagent definition for the Claude Test plugin's background spec author, which reads the app's source and writes spec drafts into a run folder's proposed/ directory for the person to approve.
Agent Prompt: Claude Test explorer (793 tks) - Subagent definition for the Claude Test plugin's read-only explorer, which maps a web application's source for a browser test suite using only Read, Grep and Glob and must avoid credential and out-of-project paths.
Agent Prompt: Claude guide agent (2975 tks) - Instructs the Claude guide agent to answer documentation-based questions about Claude Code, the Claude Agent SDK, the Claude API, Claude Tag, plugin evaluation, and skill diagnostics.
Agent Prompt: Conversation summarization (capped variant) (166 tks) - Asks for a summary of the entire conversation in 2000 words or fewer inside summary tags, preserving facts, file paths, function names, error messages and open threads.
Agent Prompt: Conversation summarization (lean variant) (486 tks) - Tells a compaction run that its reply replaces the conversation, lists what the next context window will carry, and requires the summary to preserve the user's goals, decisions, ruled-out approaches, exact limits, current status and next step.
Agent Prompt: Conversation summarization (short variant) (400 tks) - Asks for a no-tools summary block that preserves problems, options, decisions, constraints, current status, open items and exact details, weighting the user's words over Claude's own reasoning.
Agent Prompt: Dream memory consolidation (1214 tks) - Instructs an agent to perform a multi-phase memory consolidation pass — orienting on existing memories, gathering recent signal from logs and transcripts, merging updates into topic files, and pruning the index.
Agent Prompt: General purpose agent (63 tks) - Defines a general-purpose agent for researching complex questions, searching code, and executing multi-step tasks.
Agent Prompt: General purpose (446 tks) - System prompt for the general-purpose subagent that searches, analyzes, and edits code across a codebase while reporting findings concisely to the caller.
Agent Prompt: General task agent (99 tks) - Instructs a Claude Code task agent to complete the user's request fully and report the essential outcome.
Agent Prompt: Hook condition evaluator (155 tks) - Instructs an agent to decide whether a hook's user-provided check lets an action go ahead, returning an ok/reason JSON verdict.
Agent Prompt: Onboarding guide draft share link workflow (423 tks) - Adds instructions for sharing the draft ONBOARDING.md before review, then updating the same ShareOnboardingGuide link after the user answers the review questions.
Agent Prompt: Onboarding guide generator (1135 tks) - Co-authors a team onboarding guide (ONBOARDING.md) for new Claude Code users by analyzing the creator's usage data, classifying session types, and iterating on the draft collaboratively.
Agent Prompt: PR follow-up cron (PR Steward label check) (278 tks) - PR follow-up cron clause telling the agent not to fix or push a PR carrying PR Steward labels, to tell the user once, and to offer leave-it, one-change, or take-over choices.
Agent Prompt: PR follow-up cron (332 tks) - Cron prompt for checking a pull request created in the session and fixing failures, comments, or conflicts.
Agent Prompt: Plugin eval pilot trust requirement (205 tks) - Requires explicit user trust for a plugin directory before piloting its eval cases with --trust-plugin and otherwise limits work to writing case files.
Agent Prompt: Project thread status card classifier (4299 tks) - Classifies a Project thread as needing a reply, needing approval, done, failed, or working and emits concise owner-facing status-card JSON.
Agent Prompt: Pull request creation (1151 tks) - Prompt for creating a single GitHub pull request from existing commits with branch, template, attribution, shell-formatting, and git-safety guidance.
Agent Prompt: Quick PR creation (1741 tks) - Streamlined prompt for creating a commit and pull request with pre-populated context.
Agent Prompt: Quick git commit (1312 tks) - Streamlined prompt for creating a single git commit with pre-populated context.
Agent Prompt: Read-only search agent (93 tks) - Defines a read-only search agent for broad fan-out code searches that returns conclusions instead of file dumps.
Agent Prompt: Remote machine auto mode rules (228 tks) - Adds host-declared deny, allow, and environment rules to the security monitor prompt for commands that will run on another machine.
Agent Prompt: Schedule local MCP server limitation (1145 tks) - Explains that local Claude Code MCP servers cannot be attached to cloud routines and tailors Claude.ai connector guidance to why the connector list was unavailable.
Agent Prompt: Security monitor Claude Tag connector writes (542 tks) - Allows the autonomous-action security monitor to accept delegated writes through exact Claude Tag connector tool prefixes while preserving all other safety boundaries.
Agent Prompt: Security monitor attached machine call results (269 tks) - Explains how the auto mode security monitor reads _host call lines and result_from lines for calls served by a user-attached machine, and judges sending that machine's output to external destinations as cross-machine data exfiltration.
Agent Prompt: Security monitor forwarded user turns (761 tks) - Defines how the auto mode classifier authenticates harness-forwarded user turns and weighs direct, relayed, and unattributed messages as action-specific intent, consent, or restrictions.
Agent Prompt: Subagent delegation configuration classifier (474 tks) - Classifies a user's CLAUDE.md, installed skills and custom agent definitions as encouraging or not encouraging delegation to subagents, naming which sections say so while ignoring any instructions found inside the configuration.
Agent Prompt: Summarization no-tools guard (164 tks) - Shared prefix for compaction summarization agents that forbids tool use and requires a plain-text reply shaped as either an analysis block followed by a summary block or a summary block alone.
Agent Prompt: Web fetch agent usage guidance (654 tks) - Explains when and how to delegate URL reading to the built-in web-fetch agent, including binary-file trust boundaries and follow-up messaging.
Agent Prompt: Web reading specialist (905 tks) - System prompt for the built-in web-fetch agent that reads untrusted URL content with WebFetch and returns a focused, source-grounded report to its caller.
Agent Prompt: Worker fork (412 tks) - Directive injected into a forked child agent, telling it to treat the inherited transcript as reference, execute one directive directly, and report once.
Agent Prompt: Workflow subagent structured output (190 tks) - Instructs an internal workflow subagent to return its final answer by calling the StructuredOutput tool exactly once with schema-valid input.
Data
The content of various template files embedded in Claude Code.
Data: /auto-mode-setup usage (406 tks) - Defines /auto-mode-setup proposal and reviewed-apply command syntax, request ID ordering, save-target validation, and integrity requirements.
Data: Artifact MCP connector guidance (351 tks) - Explains how Artifact MCP manifests declare claude.ai, built-in, and host connectors and how to discover exact server and upstream tool names.
Data: Artifact capability verification pass (317 tks) - Requires one functional pass over a page's session-declared runtime capabilities before handing over its link, assembling the preview, database read and endpoint checks to run and what to report to the user.
Data: Artifact connector call observation requirement (305 tks) - Explains how to obtain connector argument names from the loaded tool schema, learn result shapes from safe real calls, and disclose shapes that cannot be verified before publishing.
Data: Artifact connector server naming guidance (241 tks) - Explains how connector tool prefixes map to manifest server identifiers and why in-page calls must use the resolved connector display name.
Data: Artifact host MCP server guidance (420 tks) - Explains how locally configured MCP servers may be declared as Artifact host servers and distinguishes them from Claude.ai connectors and built-in servers.
Data: Artifact publish merged-over newer version file changes notice (481 tks) - Tool-result notice appended when an artifact publish was merged over a newer live version, listing files changed, added, or removed there since the base version, that stale copies must be read again, and how many further files are not listed.
Data: Auto-compact inputs changed event schema (221 tks) - Schema description for worker-resolved auto-compaction state events used by thin clients to display the effective compaction countdown.
Data: Built-in gh stand-in api command help (1149 tks) - Usage text for Claude Code's built-in gh stand-in, which supports only gh api REST requests to GitHub hosts through the session's GitHub proxy, listing its flags and how {owner}, {repo} and {branch} placeholders are filled in.
Data: Chrome browser hints control request (207 tks) - Documents the set_chrome_browser_hints control request that tells a session which connected Claude in Chrome browser to prefer, and that the hints are session-only and never override the live relay roster.
Data: Claude Code recent changes reference (1854 tks) - Reference mapping of recently removed or renamed Claude Code commands, flags, and terms to their current replacements.
Data: Claude Platform on AWS reference (1805 tks) - Reference documentation for using the Claude Developer Platform through AWS infrastructure, including AnthropicAWS clients, required region and workspace configuration, SigV4 authentication, and short-term API keys.
Data: Claude Tag (Claude in Slack) reference (1999 tks) - Offline reference for Claude Tag, Claude Code's org-managed Slack surface, covering what it is, availability, setup, configuration, and how it differs from the earlier Claude in Slack app.
Data: Claude Test spec file format (1758 tks) - Reference for the Markdown spec files in .claude-test/specs/ — front matter keys, the three required parts, data-creating specs, and what makes a spec goal-oriented and checkable from a single end screenshot.
Data: Claude gateway landing page (392 tks) - HTML status page served at a Claude Code gateway root, showing the gateway logo, the running gateway URL, the identity-provider host, an OAuth discovery link, and the gateway version.
Data: Cloud session unverified remote branch push notice (255 tks) - Notice shown when no remote of the checkout could be read, so it could not be checked whether the current branch or detached HEAD commit exists on the remote host, telling the user to push their work to a branch and start a new cloud session from it.
Data: Code change published event schema (349 tks) - Schema description for the code_change_published system event, including provenance, idempotency, trust boundaries, and best-effort delivery.
Data: Command plugin source command field (160 tks) - Describes the command field for command-based plugin sources, including its one-line directory output contract, cache-copy lifecycle, platform shell, and re-resolution behavior.
Data: Cowork plugin MCP discovery and connection (1746 tks) - Reference guidance for finding MCP connectors during plugin customization, using search and suggestion tools, mapping categories to keywords, and writing .mcp.json entries.
Data: Cowork plugin examples (3158 tks) - Reference examples of minimal, medium, and complex Cowork plugin structures with plugin metadata, skills, agents, hooks, MCP config, README, and connectors.
Data: Cross-session inbound setting (175 tks) - Describes crossSessionInbound delivery, hold, refusal, and permission-mode parity behavior for peer messages.
Data: Data visualization anti-patterns (2339 tks) - Reference list of chart and dashboard anti-patterns to check against before shipping a data visualization.
Data: Data visualization choosing a form (1177 tks) - Reference guidance for choosing the appropriate chart, stat tile, or non-chart form based on the data's job.
Data: Data visualization components (836 tks) - Reference specification for the HTML/SVG pieces, layers, and system parameters that make up a data visualization.
Data: Data visualization interaction (1318 tks) - Reference guidance for chart hover layers, tooltips, filters, linked highlighting, and interaction behavior.
Data: Dialog expiry setting (204 tks) - Describes dialogExpiry deadlines for remote permission dialogs and held cross-session messages, including defaults and overrides.
Data: DirectoryAdded hook description (216 tks) - Describes when the DirectoryAdded hook fires, its input fields, and how failures and output are handled for add-dir and register_repo_root sources.
Data: Files API reference — C# (378 tks) - C# Files API reference including beta file upload and use with beta message types.
Data: Files API reference — Go (413 tks) - Go Files API reference including file upload, listing, deletion, and usage in messages.
Data: Files API reference — Java (492 tks) - Java Files API reference including file upload, beta message references, listing, deletion, download, and metadata retrieval.
Data: Files API reference — PHP (186 tks) - PHP Files API reference with a beta-era upload example and a warning that current SDKs use the stable Files API shape.
Data: Files API reference — Python (1719 tks) - Python Files API reference including file upload, listing, deletion, and usage in messages.
Data: Gateway device code entry page (261 tks) - HTML verification page served at a gateway device endpoint, prompting the user to enter the short device code shown by Claude Code so they can sign in through their company identity provider.
Data: Gateway ignored X-Forwarded-For warning (254 tks) - Gateway startup warning that an X-Forwarded-For header was ignored because the requesting address is not covered by listen.trusted_proxies, and what to configure when a load balancer or ingress fronts the gateway.
Data: Get task output control request (201 tks) - Schema description for the get_task_output control request that reads the last 8 KiB of a background shell or Monitor task's output without a model turn, and when it is refused.
Data: Hook classifier context field (654 tks) - Schema description for the classifierContext hook response field, including trust semantics, size and timing limits, transcript visibility, and rewrite-integrity requirements.
Data: Hook mcp_server field (312 tks) - Describes the mcp_server object attached to tool-related hook payloads — the server's config key and the source its definition came from — and why trust must key on source rather than on the server name or tool-name prefix.
Data: Interrupt cancel queued parameter (646 tks) - Schema description for the optional interrupt cancel_queued request parameter and its queued-command cancellation semantics.
Data: Interrupt receipt cancelled field (291 tks) - Schema description for the cancelled UUID list returned when an interrupt request cancels queued commands.
Data: Interrupt send_now message_uuid parameter (246 tks) - Schema description for the interrupt message_uuid parameter read only beside send_now, which restricts the request to one exact-match waiting user message and answers nothing_waiting when that message is not waiting.
Data: Interrupt send_now parameter (596 tks) - Schema description for the optional interrupt send_now request parameter, which asks that waiting user messages reach the model now without acting as a Stop, and how the CLI chooses between moving the turn's waits to the background and aborting it.
Data: Interrupt worker_epoch parameter (381 tks) - Schema description for the interrupt request's worker_epoch member, which binds a service-sent interrupt to one worker life and defines the stale_worker_epoch and invalid_worker_epoch refusals.
Data: Knowledge MCP search strategies (447 tks) - Reference query patterns for using knowledge MCPs to discover organization-specific tool names, project identifiers, team names, and workflow details during plugin customization.
Data: MCP HTTP client capability validation rationale (189 tks) - Explains where required client capabilities are validated in the modern MCP HTTP request ladder and how the pre-dispatch gate affects observable precedence.
Data: MCP read resource control request (397 tks) - Documents the mcp_read_resource control request that reads one MCP Apps (SEP-1865) ui:// resource from a connected MCP server, its sandboxed-render requirement, SDK-server rejection, refused lanes and error cases.
Data: MCP server status error_code field (200 tks) - Schema description for the MCP server status error_code field, naming the host-fixable failure causes — a rejected claude.ai connector login, a rejected first-party Anthropic credential, and a project-scoped server awaiting approval — and noting that it is absent for every other failure and that new values are additive.
Data: Managed Agents quickstart — Bug hunter (2633 tks) - Bundled Managed Agents quickstart template for a bug hunter agent that splits a codebase into slices, has parallel readers and second-look reviewers find suspicions, and reports only bugs a separate prover agent demonstrated, with its agent.md definition, environment.yaml and session.yaml.
Data: Managed Agents quickstart — Contract tracker (689 tks) - Bundled Managed Agents quickstart template for a contract tracker agent that reads a Box contract, extracts clauses and dates, and creates Asana deadline and obligation tasks, with its agent.md definition and system prompt.
Data: Managed Agents quickstart — Data analyst (579 tks) - Bundled Managed Agents quickstart template for a data analyst agent that loads, cleans and charts datasets and can query Amplitude, with its agent.md definition and system prompt.
Data: Managed Agents quickstart — Deep researcher (798 tks) - Bundled Managed Agents quickstart template for a deep research agent that decomposes a question, cites primary sources and ends with a confidence-and-gaps section, with its agent.md definition and a user.define_outcome rubric.
Data: Managed Agents quickstart — Explainer video maker (1653 tks) - Bundled Managed Agents quickstart template for an explainer video maker agent that storyboards a topic, builds one scene per agent plus generated music, and joins them with ffmpeg into a captioned video, with its agent.md definition, environment.yaml and session.yaml.
Data: Managed Agents quickstart — Field monitor (813 tks) - Bundled Managed Agents quickstart template for a field monitor agent that scans technical blogs and writes a weekly Notion digest, with its agent.md definition and a Monday-morning cron deployment.
Data: Managed Agents quickstart — Incident commander (684 tks) - Bundled Managed Agents quickstart template for an incident commander agent that triages a Sentry alert, opens a Linear ticket and runs the Slack war room, with its agent.md definition and system prompt.
Data: Managed Agents quickstart — Sprint retro facilitator (870 tks) - Bundled Managed Agents quickstart template for a sprint retro facilitator agent that pulls a closed Linear sprint, reads Slack sentiment and writes a retro doc, with its agent.md definition and a twice-monthly cron deployment.
Data: Managed Agents quickstart — Structured extractor (545 tks) - Bundled Managed Agents quickstart template for a structured extractor agent that parses unstructured text into JSON that validates against a target schema, with its agent.md definition and system prompt.
Data: Managed Agents quickstart — Support agent (532 tks) - Bundled Managed Agents quickstart template for a support agent that answers from Notion docs with quoted sources and escalates low-confidence questions to Slack, with its agent.md definition and system prompt.
The definitive curated list of Claude Code resources: tools, workflows, tricks, MCP servers, and community projects. Community-maintained with 21k+ stars.
serpro69's opinionated starter config for Claude Code — batteries-included CLAUDE.md with git conventions, commit message format, test requirements, and style guides.
Piebald.