Files
My-Brain-Is-Full-Crew/references/agents-registry.md
Giacomo Nunziati 53b605379e feat: multi-platform adapter architecture (Claude Code, Gemini CLI, OpenCode) (#32)
* Fix istall/update scripts

* test: capture pre-refactor install snapshot for regression

Adds take-snapshot.sh script and the resulting snapshot/ directory,
capturing the exact vault state produced by launchme.sh before the
framework-agnosticity refactor begins.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Summary: Refactor agents/skills/hooks/mcp in agentic-platform-agnostic templates.

refactor: rename source CLAUDE.md → DISPATCHER.md (framework-neutral)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

refactor: convert agent frontmatter from tools: to neutral capabilities:

Replace Claude Code-specific `tools:` frontmatter with framework-agnostic
`mode: subagent` and `capabilities: [...]` in all 8 agent files.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

refactor: add neutral hook trigger manifests (.hook.yaml)

refactor: hooks read neutral JSON schema (args.* instead of tool_input.*)

refactor: convert .mcp.json to neutral mcp/servers.yaml

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Implement agentic-platform adapters skeleton.

build: add adapters/lib.sh skeleton with vocabulary constants

test: bash test runner for adapter helpers

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): parse_frontmatter helper with tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): parse_capabilities helper with tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): should_include helper with tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): parse_hook_yaml helper with tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): agent_body helper with tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(adapters): enumerate_agents and enumerate_hooks helpers

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Implement agentic-platform adapter for Claude Code.

build(adapters): claude-code adapter skeleton with capability/event tables

build(claude-code): adapter_translate_dispatcher with test

build(claude-code): adapter_translate_references with test

build(claude-code): adapter_translate_skills with tests

build(claude-code): adapter_translate_agents with capability→tools mapping

build(claude-code): hook wrapper template (CC native → neutral schema)

build(claude-code): adapter_translate_hooks with wrapper generation

build(claude-code): adapter_translate_mcp with hand-rolled YAML parser

build(claude-code): adapter_finalize and complete adapter_build wiring

build: scripts/build.sh dispatches to per-framework adapter

Also fix adapter_translate_hooks and adapter_translate_agents to use
while-read loops (avoiding word-splitting on paths with spaces) and
guard grep calls with || true to survive set -eo pipefail when hooks
have no match-tool field. Remove scripts/build.sh from .gitignore.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Refactor install/update scripts to support agentic-platform agnosticity.

refactor(lib.sh): generalize install_claude_md → install_dispatcher

New signature takes the full destination path instead of just the vault
dir, allowing callers to install CLAUDE.md, AGENTS.md, or any dispatcher
file to an explicit location.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(launchme): support --framework flag, build dist/ before install

Add --framework and --target arg parsing. Run build.sh before installing
to populate dist/<framework>/. All install_* calls now read from
dist/<framework>/ instead of the raw source dirs. MCP is now handled
automatically by the adapter (no interactive prompt). Replaced
install_claude_md with install_dispatcher.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(updateme): support --framework flag, build dist/ before update

Add --framework and --target arg parsing. Run build.sh before installing
to populate dist/<framework>/. All install_* calls now read from
dist/<framework>/ instead of raw source dirs. Replaced install_claude_md
with install_dispatcher.

Also fix set -e compatibility in lib.sh: add || true to all conditional
[[ ... ]] && info "..." logging lines so they don't abort the script
when VERBOSE_COPY=0.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test: regression runner diffs dist/claude-code against pre-refactor snapshot

- Add tests/regression/run.sh that builds dist/claude-code and compares
  against snapshot, excluding runtime-only artifacts (.mbifc-manifest,
  .mcp.json, .claude-plugin/plugin.json)
- Fix adapters/lib.sh agent_body: preserve '---' section dividers in body
  (awk now only skips '---' while still inside frontmatter, fm < 2)
- Fix adapters/claude-code/adapter.sh: change 'read' capability to expand
  to only 'Read', appending 'Glob, Grep' at end of tools list to match
  snapshot ordering
- Update snapshot to reflect intentional refactor changes: hook JSON schema
  (.args.* instead of .tool_input.*), wrapper scripts, settings.json with
  wrapper paths, and consistent tool ordering for postman/sorter

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Implement opencode adapter.

Co-Authored-By: win0na <winnie@winneon.moe>

feat(lib.sh): add install_plugins helper for opencode JS plugins

build(adapters): opencode adapter skeleton with capability/event tables

build(opencode): adapter_translate_dispatcher (DISPATCHER.md → AGENTS.md)

build(opencode): adapter_translate_references and adapter_translate_skills

Implements Task 4 and Task 5:
- adapter_translate_references: Copies reference markdown files to .opencode/references/
- adapter_translate_skills: Copies skill SKILL.md files to .opencode/skills/<name>/ with exclude filtering

Both functions respect framework filtering via should_include().

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(opencode): adapter_translate_agents with capability→permission mapping

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(opencode): bash-executor template for spawning hook scripts

build(opencode): plugin-stub template for mbifc-hooks.js

build(opencode): adapter_translate_hooks with JS plugin generation

Implements _oc_hook_registry_json and adapter_translate_hooks in the
opencode adapter. Copies hook scripts to .opencode/hooks/, generates a
single .opencode/plugins/mbifc-hooks.js by inlining bash-executor.js and
synthesising a hook registry from *.hook.yaml files. Uses python3 for
template substitution to safely handle multi-line JS content. Adds 3
unit tests (copies scripts, registry entries, noop when no hooks dir).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(opencode): adapter_translate_mcp with local/remote handling

build(opencode): adapter_finalize and complete adapter_build wiring

Add adapter_finalize placeholder and wire adapter_translate_mcp into
adapter_build; add end-to-end integration test (14/14 pass).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(launchme): branch on --framework for opencode install layout

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(updateme): branch on --framework for opencode install layout

Mirror the same case "$FRAMEWORK" block from launchme.sh: framework-specific
DIST_COMPONENTS_DIR, VAULT_COMPONENTS_DIR, DISPATCHER_SRC/DST, MCP_SRC/DST,
HAS_PLUGINS; conditional install_plugins; conditional install_settings;
framework-aware vault-setup check; framework-neutral summary messages.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Fix adapters to follow the same template.

fix: restore adapter_build() contract, revert function renames

Both adapters now export adapter_build() and adapter_translate_*() as
the uniform public contract. scripts/build.sh sources one adapter and
calls adapter_build uniformly. Private helpers (_oc_*) and vocabulary
tables (cc_capability_to_tools, oc_capability_to_permission, etc.)
retain their prefixes. CC regression and OC unit tests all pass.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

fix(tests): restore test_oc_ prefix on adapter_build end-to-end test

* Fix agent format in opencode adapter

* refactor: rename --framework to --platform across all scripts and tests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Modify generic name for model tiers

Co-Authored-By: win0na <winnie@winneon.moe>

refactor: neutral model vocabulary (low/mid/high) in source agents

feat(claude-code): cc_model_to_native() maps low/mid/high to haiku/sonnet/opus

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(opencode): update oc_model_to_provider() for low/mid/high vocabulary

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Add gemini-cli adapter

Co-Authored-By: win0na <winnie@winneon.moe>

build(gemini-cli): adapter skeleton with capability/event/model tables

build(gemini-cli): adapter_translate_dispatcher (DISPATCHER.md → GEMINI.md)

build(gemini-cli): adapter_translate_references and adapter_translate_skills

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(gemini-cli): adapter_translate_agents with capability→tools mapping

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(gemini-cli): adapter_translate_hooks with wrapper scripts

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

feat(install): add gemini-cli platform to launchme.sh and updateme.sh

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Implement preserving config merge for opencode.

Co-Authored-By: win0na <winnie@winneon.moe>

feat(opencode): config-merge.sh with formatting-preserving JSON merge

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

build(opencode): source config-merge.sh from adapter

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat(install): use oc_config_merge for opencode.json instead of overwrite

Source config-merge.sh from install scripts for opencode platform so
user keys in opencode.json are preserved on reinstall and update.
Fix in-place merge by writing to a temp file before moving to output.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Add mcp files to gitignore.

* Fix claude-specific references in agents, skills and references

* Fix: remove claude-specific reference from hooks.

build: add platform_dir and dispatcher_name to all hook wrapper/plugin templates

feat(hooks): platform-aware path checks using platform_dir and dispatcher_name from JSON input

test: update regression snapshot for platform-aware hook wrappers and scripts

* Added interactive platform choice in launchme, and platform auto-detection in updateme.

* Fix: remove claude-specific references from documentation

* Update documentation to reflect the new platform-agnostic architecture

* fix: address Copilot review feedback on PR #32

- tests/run.sh: check source return code, report failures
- tests/regression/run.sh: use mktemp + trap cleanup instead of fixed /tmp paths
- tests/regression/run.sh: include .mcp.json in regression comparison
- tests/regression/take-snapshot.sh: use --platform flag instead of stale scripted input
- adapters/opencode/templates/plugin-stub.js.tmpl: include stdout in hook block error message

* fix: address Copilot review round 2

- config-merge.sh: reword comment to only promise indentation preservation (not full formatting)
- take-snapshot.sh: copy required artifacts explicitly, optional ones with existence check
- adapters/lib.sh: document parse_hook_yaml single-trigger limitation

* fix: address Copilot review round 3

- adapters/opencode/adapter.sh: replace python3 template substitution with
  pure bash (while-read loop with case matching), removing python3 dependency
- adapters/lib.sh: should_include now falls back to plain YAML key read for
  files without frontmatter delimiters (fixes hook .yaml exclude: support)

* fix: address Copilot review round 4

- scripts/launchme.sh: fix double-dot in FW_DIR_NAME display (basename
  already includes the dot, e.g. ".claude")
- scripts/launchme.sh: replace undefined MCP_ANSWER with check on MCP_DST
  existence for summary banner

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-10 23:28:28 +02:00

7.9 KiB
Executable File

Agent Registry

This file is the single source of truth for all active agents in the crew. The dispatcher (DISPATCHER.md) and all agents reference this file for routing decisions and inter-agent coordination.

The registry is designed to grow: custom agents (see Issue #12) are added as new rows following the same schema.


Registry

Name Role Capabilities Input Output Status
architect Vault Structure & Governance Create/modify folders, templates, MOCs, tag taxonomy, naming conventions. Full Bash access. Runs onboarding. Vault setup, new areas/projects, structural changes, defrag, onboarding Folders created, templates defined, structure updated, MOCs generated active
scribe Text Capture & Refinement Create notes in 00-Inbox/, format raw text, handle voice-to-note, brainstorm, quotes, reading notes Raw text, ideas, thoughts, voice input, quotes, brainstorm requests Structured notes in 00-Inbox/ with frontmatter, tags, suggested connections active
sorter Inbox Triage & Filing Move notes from inbox to correct locations, update MOCs, batch processing Inbox triage, filing requests, note organization Notes moved to correct folders, MOCs updated, triage reports active
seeker Search & Intelligence Full-text search, metadata queries, relationship navigation, answer synthesis. Read-only by default. Search queries, "find X", "where did I put", factual questions about vault content Search results with citations, synthesized answers, knowledge gap reports active
connector Knowledge Graph & Link Analysis Add/edit wikilinks, analyze graph structure, discover connections, bridge notes Link analysis, "find connections", graph health, serendipity requests New wikilinks added, graph health score, connection maps, bridge notes active
librarian Vault Health & Quality Assurance Detect/merge duplicates, fix broken links, audit frontmatter, growth analytics. Full Bash access. Maintenance, audit, cleanup, health check, duplicate detection Health reports, fixed links, merged duplicates, consistency reports active
transcriber Audio & Meeting Intelligence Process transcriptions into structured notes, extract action items, speaker detection Audio recordings, transcriptions, meeting notes, lecture/podcast processing Structured meeting/lecture notes in 00-Inbox/ with action items, decisions, topics active
postman Email & Calendar Intelligence Read/archive/delete email (Gmail via gws, Hey.com via hey), search emails, read/create/update calendar events, draft and send replies. Uses Google Workspace CLI (gws) and/or Hey CLI (hey) via Bash, with MCP as read-only fallback. Email triage, calendar queries, deadline tracking, meeting prep, VIP filtering Email summaries saved as notes in 00-Inbox/, calendar events created, deadline reports active

Status Values

  • active: Agent is operational and available for dispatch
  • disabled: Agent is temporarily disabled — the dispatcher will skip it

How This File Is Used

  1. Dispatcher reads the Input column to match user messages to agents
  2. Dispatcher reads Output + Capabilities of other agents to decide if chaining is needed after an agent returns
  3. Agents reference this file when suggesting next agents in their output
  4. Custom agents are added as new rows by the Architect during the custom agent creation flow

Custom Agents

Custom agents are created by the Architect through a conversational flow with the user. They follow the exact same schema as core agents and are added as new rows in the Registry table above.

How Custom Agents Are Added

  1. The user asks the Architect to create a new agent (or an existing agent suggests one via ### Suggested new agent)
  2. The Architect conducts a detailed conversation to understand requirements
  3. The Architect generates the agent file in .platform/agents/, adds a row to the Registry table above, and updates agents.md
  4. The platform auto-discovers the new agent from its frontmatter

Naming Rules

  • Custom agent names must be lowercase, hyphens only (e.g., habit-tracker, recipe-manager)
  • Names must NOT conflict with core agent names: architect, scribe, sorter, seeker, connector, librarian, transcriber, postman
  • Names should be descriptive and concise (1-2 words)

Priority

Custom agents always have lower routing priority than the 8 core agents. The dispatcher checks custom agents only when no core agent matches the user's message. Among custom agents, the dispatcher uses the Input column to find the best match


Skills Registry

Skills handle complex, multi-step workflows extracted from agents. They are checked before agents by the dispatcher (higher priority). Skills run in the main conversation context via the Skill tool, preserving multi-turn state.

Skill Source Agent Triggers Purpose Status
/onboarding architect "initialize the vault", "set up the vault", "onboarding", "vault setup" Full vault setup conversation active
/create-agent architect "create a new agent", "custom agent", "I need a new agent", "build an agent", "new crew member" Custom agent creation (6-phase interview) active
/manage-agent architect "edit my agent", "update agent", "remove agent", "delete agent", "list agents", "show my agents" Edit, remove, list custom agents active
/defrag architect "defragment the vault", "reorganize the vault", "structural maintenance", "vault defrag", "weekly defrag" Weekly vault defragmentation (5-phase audit) active
/email-triage postman "check my email", "what's in my inbox", "process emails", "email triage", "anything urgent in email?" Email scanning, priority scoring, classification active
/meeting-prep postman "prepare for meeting", "meeting prep", "brief me for the meeting", "get ready for the call" Comprehensive meeting brief with context gathering active
/weekly-agenda postman "weekly agenda", "what's this week", "week overview", "plan my week" Day-by-day week overview from calendar, email, vault active
/deadline-radar postman "deadline radar", "what are my deadlines", "this week's deadlines", "upcoming deadlines" Unified deadline timeline with urgency grouping active
/transcribe transcriber "transcribe", "I have a recording", "process this audio", "meeting notes from recording", "summarize the call" Audio/transcript processing with structured notes active
/vault-audit librarian "weekly review", "check the vault", "vault audit", "full audit", "vault health" Full 7-phase vault audit active
/deep-clean librarian "deep clean", "deep cleanup", "thorough cleanup", "the vault is a mess" Extended vault cleanup with stale content detection active
/tag-garden librarian "tag garden", "clean up tags", "tag cleanup", "tag audit" Tag analysis: unused, orphan, near-duplicates active
/inbox-triage sorter "triage the inbox", "clean up the inbox", "sort my notes", "empty inbox", "file my notes", "process the inbox" Inbox note processing, classification, and routing active
/contact-sync postman "sync contact", "add to contacts", "save contact", "update contact", "is this person in my contacts" Sync person to Apple Contacts (search, create, update). Requires apple-contacts MCP. active

How Skills Are Routed

  1. The dispatcher checks the skill routing table (in DISPATCHER.md) before the agent routing table
  2. If a trigger matches, the skill is invoked via the Skill tool — not the Agent tool
  3. If no skill matches, the dispatcher falls through to agent routing
  4. Skills can produce ### Suggested next agent output, which the dispatcher handles using the same chaining rules as agents