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 + tier split + 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".
What this skill does
# 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`](../../templates/claude-md-doctrine.md). Doctrine background: [DOCTRINE.md](../hyperflow/DOCTRINE.md). ## Subcommands | Subcommand | Description | |---|---| | `generate` | Write the doctrine block into the project's `CLAUDE.md` (create the file if absent, append the block if not present, refresh if already present) | | `refresh` | Same as `generate` — alias for clarity when the block already exists | | `remove` | Remove 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) | | `status` | Show 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`: | Mode | Behavior | |---|---| | `auto` (**default** when `.bridge-mode` is absent) | If `./CLAUDE.md` is missing the doctrine block OR has an outdated version, **silently writes/refreshes** the block and prints a one-line notice in session-start output.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. | | `manual` | Never 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 Prints a one-line advisory when the block is missing or outdated: `./CLAUDE.md doctrine block would be refreshed (version 4.11.0) — run /hyperflow:bridge refresh to apply`. | | `off` | Does 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, tier split, 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 | Situation | Use 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 / Windows | **Yes** — bridge gives Desktop the autonomy + intent-routing + commit cadence rules | | You use claude.ai web app for this project | **Yes** — same reason | | You use VS Code / Cursor / JetBrains and the extension shells out to the `claude` CLI | No — the CLI plugin applies | | You use VS Code / Cursor and the extension talks to the API directly | **Yes** — the API session loads `CLAUDE.md` | | You collaborate with teammates who use mixed surfaces | **Yes** — commit the generated `CLAUDE.md` so everyone has the same rules regardless of their surface | ## What you keep / lose vs the full CLI plugin | Capability | CLI plugin | CLAUDE.md bridge | |---|---|---| | Autonomy rules (no confirmations, minimal output, no hedging) | yes | **yes** | | Intent-based routing (audit/debug/fix/brainstorm verbs) | yes | **yes (described in CLAUDE.md as rules for the orchestrator to follow)** | | Per-task commit cadence | yes | **yes** | | Tier split (per-batch Sonnet, final Opus) | yes | **yes** | | File-first artefacts under `.hyperflow/` | yes | **yes** | | Binary-gate rule (no recommendation on yes/no) | yes | **yes** | | No-AI-attribution rule | yes | **yes** | | Security blocklists | yes | **yes** | | `/hyperflow:*` slash commands | yes | no — surfaces without the plugin can't dispatch named skills | | Chain-mode Step-0 auto/manual question | yes | no — defaults to auto-style chain in CLAUDE.md mode | | Operational pre-elections (commit/branch/push at scope Step 2.6) | yes | no — defaults applied per CLAUDE.md guidance | | Per-step Worker → Reviewer dispatch templates from `worker-prompt.md` / `reviewer-prompt.md` | yes | partial — tier-split rule preserves the spirit; exact prompts not embedded (would bloat CLAUDE.md) | | Background agents, sticky mode, status skill, cache skill | yes | no — these need their own slash command surfaces | | Adaptive flow profiles (`fast` / `standard` / `deep`) | yes | no — 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 · Tier split (per-batch Sonnet, final Opus) · 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
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.