Bridge logo

Bridge

CommunityPopular
jeremylongshore
bridge

Use when the user wants hyperflow's behavioral rules to apply outside the terminal CLI — in Claude Code Desktop, claude.ai web, or IDE extensions that don't load CLI plugins. Writes a managed doctrine block into the project's CLAUDE.md so autonomy + intent-routing + commit cadence + role separation + file-first rules carry over. Lossy (no slash commands, no actual skill dispatch) but useful. Trigger with /hyperflow:bridge, "make hyperflow work in desktop", "make hyperflow work in claude.ai", "embed hyperflow doctrine in CLAUDE.md", "portable hyperflow rules".

Overview

Publisherjeremylongshore
Repositorytons-of-skills-marketplace
Skill namebridge
Stars
2.8K
Forks
402
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by jeremylongshore on GitHub. Read the source before you install it.

Installation

Install the Bridge AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/jeremylongshore/tons-of-skills-marketplace.git /tmp/tons-of-skills-marketplace
mkdir -p .claude/skills
cp -r /tmp/tons-of-skills-marketplace/plugins/ai-agency/hyperflow/skills/bridge .claude/skills/bridge
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Bridge in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Bridge on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Bridge is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Bridge

Embed the portable subset of hyperflow's doctrine into the project's CLAUDE.md so it applies in surfaces that don't load CLI plugins (Claude Code Desktop, claude.ai web, IDE extensions). The doctrine block is managed via fenced markers, so refreshing it on plugin updates is idempotent and never touches your own CLAUDE.md content.

Source template: templates/claude-md-doctrine.md. Doctrine background: DOCTRINE.md.

Subcommands

SubcommandDescription
generateWrite the doctrine block into the project's CLAUDE.md (create the file if absent, append the block if not present, refresh if already present)
refreshSame as generate — alias for clarity when the block already exists
removeRemove the doctrine block from CLAUDE.md (preserves your own content; if the file becomes empty after removal, leave it as an empty file rather than delete)
statusShow whether the doctrine block is present, its version, when it was generated
mode <auto|manual|off>Set the auto-bridge mode for this project. Writes .hyperflow/.bridge-mode. The session-start hook reads this and decides what to do

Default subcommand when none provided: status.

Auto-bridge (default ON)

The CLI session-start hook (hooks/session-start) runs scripts/auto-bridge.py on every session start. Behavior depends on the mode stored in .hyperflow/.bridge-mode:

ModeBehavior
auto (default when .bridge-mode is absent)If ./CLAUDE.md is missing the doctrine block OR has an outdated version, silently writes/refreshes the block and prints a one-line notice in session-start output. Zero user friction.
manualNever writes. Prints a one-line advisory when the block is missing or outdated: ./CLAUDE.md doctrine block would be refreshed (version 4.11.0) — run /hyperflow:bridge refresh to apply.
offDoes nothing. No writes, no advisories.

This means: open Claude Code CLI once in your project, and from then on every Desktop / web / IDE session in the same project automatically gets the up-to-date hyperflow doctrine via CLAUDE.md. Refresh on plugin update is automatic too.

To opt out: /hyperflow:bridge mode off. To require explicit refresh: /hyperflow:bridge mode manual.

What gets written

A fenced block in the project's ./CLAUDE.md (at the repo root, where Claude Code Desktop / web / CLI all look for it):

markdown
<!-- hyperflow:doctrine:start version=<X.Y.Z> generated=<ISO-8601> source=https://github.com/Mohammed-Abdelhady/hyperflow -->

# Hyperflow Doctrine (Portable Subset)

<the full template body  autonomy, intent-routing, commit cadence,
 role separation, file-first artefacts, no AI attribution, security
 blocklists, what's missing vs CLI>

<!-- hyperflow:doctrine:end -->

The fenced markers (hyperflow:doctrine:start / hyperflow:doctrine:end) let refresh find and replace ONLY the doctrine block, leaving everything else in your CLAUDE.md untouched. Place the block anywhere in CLAUDE.md; the bridge respects its position.

When to use

SituationUse bridge?
You work exclusively in Claude Code CLI (terminal)No — the plugin loads doctrine directly; bridge would duplicate
You use Claude Code Desktop on Mac / WindowsYes — bridge gives Desktop the autonomy + intent-routing + commit cadence rules
You use claude.ai web app for this projectYes — same reason
You use VS Code / Cursor / JetBrains and the extension shells out to the claude CLINo — the CLI plugin applies
You use VS Code / Cursor and the extension talks to the API directlyYes — the API session loads CLAUDE.md
You collaborate with teammates who use mixed surfacesYes — commit the generated CLAUDE.md so everyone has the same rules regardless of their surface

What you keep / lose vs the full CLI plugin

CapabilityCLI pluginCLAUDE.md bridge
Autonomy rules (no confirmations, minimal output, no hedging)yesyes
Intent-based routing (audit/debug/fix/brainstorm verbs)yesyes (described in CLAUDE.md as rules for the orchestrator to follow)
Per-task commit cadenceyesyes
Role separation (workers execute, reviewers review)yesyes
File-first artefacts under .hyperflow/yesyes
Binary-gate rule (no recommendation on yes/no)yesyes
No-AI-attribution ruleyesyes
Security blocklistsyesyes
/hyperflow:* slash commandsyesno — surfaces without the plugin can't dispatch named skills
Chain-mode Step-0 auto/manual questionyesno — defaults to auto-style chain in CLAUDE.md mode
Operational pre-elections (commit/branch/push at scope Step 2.6)yesno — defaults applied per CLAUDE.md guidance
Per-step Worker → Reviewer dispatch templates from worker-prompt.md / reviewer-prompt.mdyespartial — the role-separation rule preserves the spirit; exact prompts not embedded (would bloat CLAUDE.md)
Background agents, sticky mode, status skill, cache skillyesno — these need their own slash command surfaces
Adaptive flow profiles (fast / standard / deep)yesno — orchestrator infers from message complexity

Net coverage: ~70% of hyperflow's behavioral value. Slash commands and the infrastructure that wraps them are the missing 30%.

Subcommand Details

generate / refresh

  1. Read the template at ~/.claude/plugins/cache/hyperflow-marketplace/hyperflow/<version>/templates/claude-md-doctrine.md (resolve current version from the active plugin install).
  2. Substitute placeholders: __HYPERFLOW_VERSION__ → current plugin version, __GENERATED_AT__ → current UTC timestamp (ISO-8601).
  3. Read the project's ./CLAUDE.md. Three cases:
    • File absent — create ./CLAUDE.md with just the doctrine block.
    • File exists, no existing doctrine block — append the doctrine block at the end of the file (preceded by one blank line if the file doesn't already end with a blank line).
    • File exists, doctrine block present — find the <!-- hyperflow:doctrine:start … --> and <!-- hyperflow:doctrine:end --> markers, replace everything between them (inclusive of the markers) with the new block. All other content in CLAUDE.md is preserved exactly.
  4. Write the updated CLAUDE.md.
  5. Print:
Wrote hyperflow doctrine block to ./CLAUDE.md (version 4.10.1).
Surfaces that load CLAUDE.md (Desktop, claude.ai web, IDE extensions that talk to API) will now honor:
  · Autonomy rules
  · Intent-based routing (audit/debug/fix/brainstorm/scope/deploy verbs)
  · Per-task commit cadence
  · Role separation (workers execute, reviewers review)
  · File-first artefacts under .hyperflow/
  · No AI attribution
  · Security blocklists

Re-run `/hyperflow:bridge refresh` after updating the plugin to pick up doctrine changes.
What's NOT in the bridge: /hyperflow:* slash commands, plugin-loaded skill files, operational pre-elections. Those need the terminal CLI.

remove

  1. Read ./CLAUDE.md. If absent or doctrine markers not present, print Nothing to remove — no hyperflow doctrine block in ./CLAUDE.md. and stop.
  2. Find the <!-- hyperflow:doctrine:start … --> and <!-- hyperflow:doctrine:end --> markers; remove everything between them (inclusive of markers). Collapse adjacent blank lines so the file doesn't end up with a triple newline.
  3. If CLAUDE.md is now empty (or only whitespace), leave it as an empty file rather than delete — the user may have other tooling that expects the file to exist.
  4. Print Removed hyperflow doctrine block from ./CLAUDE.md. Surfaces that loaded the doctrine block will revert to default behaviour.

mode <auto|manual|off>

Write the chosen mode to .hyperflow/.bridge-mode. The file holds one word: auto, manual, or off. The session-start hook reads it on every CLI session start. Print one of:

Auto-bridge: AUTO — ./CLAUDE.md doctrine block is silently maintained on every CLI session start.
Auto-bridge: MANUAL — session start prints an advisory when the block is stale; you run /hyperflow:bridge refresh.
Auto-bridge: OFF — no advisories, no writes. Use /hyperflow:bridge generate manually if you want the block.

Defaults to auto when .bridge-mode is absent (so first install of hyperflow auto-bridges with no user action). Setting auto explicitly is a no-op write that just records the intent.

status

Read ./CLAUDE.md. Find the doctrine markers. Print one of:

Hyperflow doctrine block: PRESENT in ./CLAUDE.md
  Version generated: 4.10.1
  Generated at:      2026-05-17T15:30:00Z
  Plugin current:    4.10.1
  Status:            up to date · or · update available (re-run /hyperflow:bridge refresh)
Hyperflow doctrine block: NOT PRESENT in ./CLAUDE.md
  Use /hyperflow:bridge generate to add it.
Hyperflow doctrine block: NOT PRESENT (no ./CLAUDE.md in project root)
  Use /hyperflow:bridge generate to create ./CLAUDE.md with the doctrine block.

Flow

  1. Parse subcommand (default status).
  2. Locate the project's ./CLAUDE.md (repo root).
  3. Execute subcommand per the details above.
  4. Print one-block confirmation.

Overview

/hyperflow:bridge is the user-facing interface for the CLAUDE.md doctrine bridge. It does NOT itself enforce the doctrine — it writes the rules into a file that surfaces outside the terminal CLI will load. Enforcement happens in those surfaces when the host loads CLAUDE.md at session start.

Prerequisites

  • Project root contains a writable ./ directory (the bridge writes ./CLAUDE.md).
  • Hyperflow plugin installed (the bridge reads its template from the plugin cache). If running outside the plugin context, the bridge falls back to a vendored copy in the plugin source tree.

Instructions

See Subcommands and Subcommand Details. Summary:

  1. Parse subcommand (default status).
  2. Read/write ./CLAUDE.md per the chosen subcommand.
  3. Print one short confirmation block.

Output

  • generate / refresh — multi-line confirmation listing what surfaces now honor; what's NOT in the bridge.
  • remove — one-line confirmation.
  • status — three-field block (present / version / freshness).

Error Handling

FailureBehavior
./CLAUDE.md not writablePrint explicit error; suggest chmod +w ./CLAUDE.md if permissions-related. Do NOT silently fall back to a hidden location.
Plugin template not found (running outside cached plugin)Look up the template at <plugin-root>/templates/claude-md-doctrine.md relative to this SKILL.md's directory. If still not found, refuse with a clear error.
Existing doctrine block has malformed markers (one without the other)Refuse to refresh; print the malformed file's line range and ask the user to fix the markers manually. Do NOT auto-repair — the user's content might be at risk.
Multiple doctrine blocks in one CLAUDE.md (duplicate)Refuse and surface the line ranges of both. User decides which to keep.
User runs generate repeatedlyIdempotent — every run replaces the block with the latest template + timestamp. No duplication.

Examples

First-time setup for a Desktop user

You: /hyperflow:bridge generate

Wrote hyperflow doctrine block to ./CLAUDE.md (version 4.10.1).
Surfaces that load CLAUDE.md (Desktop, claude.ai web, IDE extensions that talk to API) will now honor:
  · Autonomy rules
  · Intent-based routing (audit/debug/fix/brainstorm/scope/deploy verbs)
  · Per-task commit cadence
  · Role separation (workers execute, reviewers review)
  · File-first artefacts under .hyperflow/
  · No AI attribution
  · Security blocklists

Re-run `/hyperflow:bridge refresh` after updating the plugin to pick up doctrine changes.
What's NOT in the bridge: /hyperflow:* slash commands, plugin-loaded skill files, operational pre-elections. Those need the terminal CLI.

Status check

You: /hyperflow:bridge status

Hyperflow doctrine block: PRESENT in ./CLAUDE.md
  Version generated: 4.10.0
  Generated at:      2026-05-15T09:12:00Z
  Plugin current:    4.10.1
  Status:            update available (re-run /hyperflow:bridge refresh)

Refresh after plugin update

You: /hyperflow:bridge refresh

Wrote hyperflow doctrine block to ./CLAUDE.md (version 4.10.1).
...

Remove

You: /hyperflow:bridge remove

Removed hyperflow doctrine block from ./CLAUDE.md. Surfaces that loaded the doctrine block will revert to default behaviour.

Resources

Frequently asked questions

What does the Bridge AI skill do?

Use when the user wants hyperflow's behavioral rules to apply outside the terminal CLI — in Claude Code Desktop, claude.ai web, or IDE extensions that don't load CLI plugins. Writes a managed doctrine block into the project's CLAUDE.md so autonomy + intent-routing + commit cadence + role separation + file-first rules carry over. Lossy (no slash commands, no actual skill dispatch) but useful. Trigger with /hyperflow:bridge, "make hyperflow work in desktop", "make hyperflow work in claude.ai", "embed hyperflow doctrine in CLAUDE.md", "portable hyperflow rules".

Why use Bridge on TypingMind?

Because you install it once and use it with any model. Bridge is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Bridge in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/jeremylongshore/tons-of-skills-marketplace/tree/main/plugins/ai-agency/hyperflow/skills/bridge. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Bridge?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Bridge?

As many as you like. As long as a model supports skills, you can use Bridge with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Bridge AI skill free?

Yes. It is published on GitHub by jeremylongshore under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇