1ee2dc415f
* Add standardized template for custom agent creation
New file: references/agent-template.md
This is a reference document that the Architect reads when generating
custom agents. It defines the exact structure every agent must follow:
YAML frontmatter format, required sections (Language, User Profile,
Inter-Agent Coordination, Core Responsibilities, Operational Rules),
placeholder tokens, and inline conventions (naming rules, tool
permissions, multilingual triggers).
The template is not an agent itself. It is a structural guide that
ensures custom agents are generated with the same quality and
consistency as the core 8.
* Add custom agent support to orchestration references
agents-registry.md:
- Added "Custom Agents" section with rules for how custom agents
are added to the registry (naming, priority, creation flow)
- Custom agents always have lower priority than core 8
- Names must be lowercase with hyphens, no conflicts with core names
agents.md:
- Added "Custom Agents" section explaining what they are, how they
coordinate with core agents, and how to create/edit/remove them
agent-orchestration.md:
- Added "Suggested new agent" signal format so agents can flag when
the user needs functionality that no existing agent provides
- Added step in Dispatcher Decision Logic to check for this signal
- Added "Custom Agent Lifecycle" section covering creation, discovery,
routing, chaining, maintenance, and deletion
* Add "Suggested new agent" capability to all 7 non-architect agents
Each agent now has a "When to suggest a new agent" subsection inside
its Inter-Agent Coordination block. When an agent detects that the
user needs functionality that no existing agent can handle, it
outputs a "Suggested new agent" section with:
- Need: what capability is missing
- Reason: why no existing agent covers it
- Suggested role: what the new agent would do
The dispatcher reads this and asks the user if they want the
Architect to create a custom agent for the detected need.
Agents are also told when NOT to suggest a new agent (existing
agent can handle it, one-off task, outside vault scope).
* Add full custom agent creation flow to the Architect
The Architect can now create, edit, and remove custom agents through
an extended conversational flow with the user.
Changes to the frontmatter:
- Added trigger phrases for custom agent creation in 6 languages
("create a new agent", "custom agent", "crea un nuovo agente", etc.)
New section "Custom Agent Creation" with:
- 5-phase conversation flow (Understanding, Capabilities, Output,
Advanced, Confirmation) where the Architect asks one question at
a time and adapts follow-ups based on user answers
- Agent file generation following references/agent-template.md
- Automatic updates to agents-registry.md and agents.md
- Management commands: edit, remove, list custom agents
- Validation rules: no name conflicts with core 8, minimal tool
permissions by default, mandatory coordination sections
- Quality standards: Core Responsibilities must be detailed enough
to produce a production-quality agent
Also added the "Suggested new agent" subsection (same as other agents).
* Update dispatcher routing to support custom agents
- Changed "The ONLY agents you may use are these 8" to acknowledge
that custom agents created by the Architect are also valid
- Added row 9+ to the routing priority table for custom agents
(always lower priority than core 8)
- Added section 9 "CUSTOM AGENTS" with routing logic: when no core
agent matches, check agents-registry.md for custom agents
- Added check for "Suggested new agent" signals in the multi-agent
routing decision flow (step 6)
* Add custom agents to README, CONTRIBUTING, and TERMS_OF_USE
README.md:
- Badge changed from "8 Agents" to "8+ Agents"
- Added custom agents as point 4 of "What makes this different",
positioned right after the main pitch for maximum visibility
- Includes a table of real-life scenarios (budget tracking, journaling,
paper reading, project monitoring, client deadline management)
- Links to Terms of Use for the responsibility disclaimer
CONTRIBUTING.md:
- Renamed "Propose a new crew member" to "Propose a new core crew
member" with a note that users can create custom agents via Architect
- Added "Custom agents vs. core agents" section explaining the
distinction between personal custom agents and project-shipped
core agents
TERMS_OF_USE.md:
- Added new Section 9 "Custom Agents" with full disclaimer: custom
agents are entirely the user's creation and responsibility, no
warranty on their behavior, author accepts no liability
- Updated Section 7 (Limitation of Liability) to reference custom
agents explicitly
- Renumbered sections 9-11 to 10-12
* Force the Architect to always run the full conversation before creating a custom agent
The Architect was generating custom agents immediately from a single
user message instead of going through the 5-phase conversation flow.
Added explicit blocking instructions at three points:
- Section intro: "NEVER create an agent in one shot"
- Before the conversation phases: "Do NOT generate the agent
immediately, even if the request seems clear"
- Explicit rule: "You are NOT allowed to create the agent file
until Phase 5"
- Reinforced one-question-per-message rule
* Force custom agent descriptions to use only the user's language
The Architect was copying the multilingual pattern from core agents
and adding translations in 6+ languages to custom agent descriptions.
Custom agents should have their description and trigger phrases
written exclusively in the language the user speaks. Reinforced
this rule in both the generation instructions and the validation
rules section.
* Force custom agent body to always be written in English
The frontmatter description uses the user's language (for trigger
matching), but the agent body (system prompt) must always be in
English for better LLM instruction-following performance. The agent
still responds in the user's language at runtime thanks to the
language matching rule.
* Add confirmation prompt before overwriting existing installation
launchme.sh:
- Detects if .claude/ or CLAUDE.md already exist in the vault
- Shows the user what will be overwritten
- Asks for explicit confirmation before proceeding
- Clarifies that custom agents and vault notes are never touched
updateme.sh:
- Asks for confirmation before overwriting core files
- Same clarification about custom agents being preserved
* Deprecate removed files instead of leaving orphans in the vault
When a file is removed from the repo (agents, references, or skills),
the updater now renames it with a "-DEPRECATED" suffix and prepends
a "DEPRECATED DO NOT USE" header instead of silently leaving it.
updateme.sh:
- Core agents in .claude/agents/ that no longer exist in the repo
get renamed to {name}-DEPRECATED.md with deprecation header
(custom agents are never touched)
- References in .claude/references/ that no longer exist in the repo
get the same treatment
- The entire .claude/skills/ directory (removed from the project)
gets renamed to .claude/skills-DEPRECATED/ with deprecation headers
on each SKILL.md
- Summary now reports deprecated file count
launchme.sh:
- Added confirmation prompt before overwriting existing installation
- Skills generation and copying kept intact (generate-skills.py runs
first, then copies to .claude/skills/)
* Add first-run setup section to custom agent template and creation flow
Custom agents now have a "First Run Setup" section that defines what
the agent must do the very first time it is invoked: what questions
to ask the user, what config files or folders to create, and how to
detect that it has already been set up.
agent-template.md:
- New "First Run Setup" section between Core Responsibilities and
Operational Rules, with subsections for detection, questions to
ask, what to create, and post-setup behavior
architect.md:
- New Phase 4 "First Run Setup" in the conversational flow where
the Architect asks the user what the agent should do on first run
- Previous Phase 4 (Advanced) becomes Phase 5
- Previous Phase 5 (Confirmation) becomes Phase 6
- Updated blocking rules to reference Phase 6
* Redesign README header and broaden legal disclaimers for custom agents
Improve README header visual hierarchy: centered title, prominent Discord
CTA, metadata badges moved to secondary row. Replace two custom agent
examples with funnier, gender-neutral ones. Rewrite TERMS_OF_USE Section 3
from "Health and Wellness Agents" to "Custom Agents and Advice-Generating
Output" covering health, legal, financial, and all regulated domains.
* Fix reference paths in architect to use .claude/references/ prefix
The agent runs inside the vault where references live under
.claude/references/, not under references/ (which is the repo layout).
* Fix reference paths in agent-template to use .claude/references/ prefix
* Add core-manifest to protect custom agents from deprecation
launchme.sh and updateme.sh now write .core-manifest listing which agent
files were installed as core. The deprecation loop checks this manifest
before touching any file, so custom agents are never deprecated.
* Skip agent deprecation if target DEPRECATED file already exists
* Include skill count in updateme.sh summary condition and message
* Fix agents-registry.md paths in CLAUDE.md to use .claude/references/ prefix
* Add edit/remove/list trigger phrases to custom agent routing
* Deprecate stale core agents on reinstall before copying new ones
On reinstall (EXISTING=1), read the old .core-manifest and deprecate
any agent that is no longer shipped in the repo, before writing the
new manifest. Prevents stale core agents from lingering in the vault.
* Fix custom agent description: triggers are in user's language, not multilingual
* Fix updateme.sh: deprecate stale agents before rewriting manifest
The manifest was being truncated and rewritten before the deprecation
loop ran, so removed core agents were no longer in the manifest and
got skipped as "custom". Now: read old manifest -> deprecate -> copy
new agents -> rewrite manifest.
* Fix grep exit code handling when removing last entry from manifest
* Fix nested fenced code blocks in agent-template using tildes
* Add core-manifest for references to protect user-created reference docs
Same pattern as agents: launchme/updateme write a .core-manifest in
.claude/references/ listing installed core files. The deprecation loop
only touches files in the manifest, leaving user-created references
untouched.
* Move deprecated files to .claude/deprecated/ to prevent auto-discovery
Deprecated agents kept in .claude/agents/ could still be auto-discovered
by Claude Code via their frontmatter. Moving them to .claude/deprecated/
ensures they are completely invisible to the dispatcher while still
preserved for user reference.
* Harden updateme.sh from Copilot review feedback
Address multiple issues raised during PR code review:
- Skip deprecation entirely when .core-manifest is missing, preventing
accidental deprecation of custom agents/references on first update
- Preserve user's "## Custom Agents" sections in agents-registry.md and
agents.md during reference updates (merge strategy instead of overwrite)
- Update confirmation message to accurately reflect what is preserved
* Preserve custom agent content during install and update
- launchme.sh: skip overwriting agents-registry.md and agents.md on
reinstall to preserve custom agent entries
- updateme.sh: extract and re-insert custom table rows from the
registry table plus custom sections, preventing data loss when
updating from upstream
- Require manifest before deprecating to avoid false positives
* Update wardrobe-coach example phrase in README
* Use robust string matching and printf for user-mutable refs
- Replace grep -qw with bash substring match for filename detection
- Replace echo with printf '%s\n' to prevent content mangling
* Enforce step-by-step conversation in Architect and fix registry row reinsertion
- Add HARD CONSTRAINT blocks to both onboarding and custom agent creation
flows, forcing the use of AskUserQuestion for each question to prevent
the Architect from skipping phases or bundling questions
- Replace hard-coded "| postman |" match in updateme.sh with generic
last-table-row detection to avoid breaking custom row reinsertion if
core agents are renamed or reordered
* Improve input handling in launchme.sh and updateme.sh for non-interactive shells
* Extract 13 skills from agents and update full documentation
Architecture change: complex multi-step flows (onboarding, email triage,
transcription, etc.) are now skills that run in the main conversation
context instead of agent subprocesses. This fixes the state/context loss
that caused agents to skip phases during multi-turn conversations.
Skills created (13):
- Architect: /onboarding, /create-agent, /manage-agent, /defrag
- Postman: /email-triage, /meeting-prep, /weekly-agenda, /deadline-radar
- Transcriber: /transcribe
- Librarian: /vault-audit, /deep-clean, /tag-garden
- Sorter: /inbox-triage
Agent changes:
- architect.md: -70% (1554 → 473 lines)
- transcriber.md: -72% (530 → 147 lines)
- postman.md: -42%, librarian.md: -37%, sorter.md: -11%
- All agents: explicit post-it create-if-not-exists
Scripts:
- launchme.sh/updateme.sh: copy skills/ directly, remove generate-skills.py
Docs updated:
- README.md: new Skills section, mermaid diagrams, routing
- getting-started.md, examples.md: skill references
- docs/agents/*.md: capability tables with skill vs agent routing
- references/agents.md, agent-orchestration.md, agents-registry.md,
agent-template.md: skill registry, skill-first routing protocol
15 KiB
Terms of Use
My Brain Is Full - Crew Last updated: 2026-03-23
By using this software, you agree to the following terms. If you do not agree, do not use the software.
1. Nature of the Software
This software is a collection of AI agent prompts ("the Crew") designed to help individuals organize personal notes, tasks, and information inside an Obsidian vault using Claude Code. It is an open-source tool provided free of charge under the MIT License.
The software does not collect, transmit, or store any data outside your local device. All data remains in your Obsidian vault on your filesystem.
2. Personal Use Only
This software is designed and intended exclusively for personal, non-commercial use on your own data. It is not designed for:
- Professional or commercial data processing
- Use by organizations, companies, or teams on behalf of third parties
- Processing of data belonging to other individuals without their knowledge and consent
- Any use that would qualify you as a "data controller" or "data processor" under applicable law without the appropriate legal basis and safeguards in place
If you choose to use this software outside the scope of personal use, you do so entirely at your own risk and responsibility.
3. Custom Agents and Advice-Generating Output
This software allows users to create custom agents (see Section 9) that may generate output in domains including, but not limited to: health, nutrition, fitness, diet, mental wellness, legal matters, financial planning, tax strategy, investment, career guidance, parenting, relationship advice, and any other area where professional expertise would normally be required.
By using any agent — core or custom — that produces advice-like output, you explicitly acknowledge and accept that:
a) All agents produce AI-generated output only. They are not medical devices, medical software, health apps, therapeutic tools, legal counsel systems, financial advisory platforms, or professional service tools within the meaning of any applicable regulation or professional licensing framework.
b) The output of any agent does not constitute medical advice, nutritional counseling, dietary prescription, psychological therapy, psychiatric treatment, legal advice, legal opinion, financial advice, investment recommendation, tax guidance, or any other form of professional counsel or service, regardless of how specific, detailed, or authoritative the output may appear.
c) The author of this software is not a healthcare professional, licensed therapist, registered dietitian, medical practitioner, attorney, licensed financial advisor, certified public accountant, or any other licensed or regulated professional of any kind. The author provides prompt engineering only.
d) All output that resembles professional advice — including but not limited to calorie calculations, TDEE estimates, meal plans, dietary suggestions, therapeutic techniques, mood assessments, legal interpretations, contract analyses, financial projections, investment strategies, tax calculations, and any other domain-specific guidance — is approximate, potentially inaccurate, potentially outdated, and not validated by any medical, legal, financial, scientific, or regulatory authority.
e) You must not rely on any agent as your sole or primary source of information in any domain where professional expertise is normally required. You must consult qualified, licensed professionals for all health-related, legal, financial, and other regulated decisions.
f) If you have or suspect you have a medical condition, eating disorder, mental health crisis, or any health emergency, you must immediately contact a qualified healthcare provider or emergency service. If you face a legal emergency or time-sensitive legal matter, you must immediately consult a licensed attorney in your jurisdiction. These agents are not equipped to handle emergencies of any kind.
g) You voluntarily assume all risk associated with following any suggestion, plan, technique, interpretation, or recommendation generated by any agent, whether core or custom.
h) The author accepts no liability whatsoever for any harm, injury, illness, deterioration of health, psychological distress, financial loss, legal consequence, regulatory penalty, missed deadline, or any other adverse outcome resulting from your use of any agent's output, whether directly or indirectly.
i) Custom agents are entirely user-created. The author has no control over, knowledge of, or responsibility for the prompts, domains, or behaviors of custom agents. Any custom agent that produces advice-like output is subject to all the disclaimers in this section, regardless of whether the user explicitly acknowledges it at creation time.
4. Data Protection and Privacy (GDPR, CCPA, and Applicable Laws)
4.1. Your Responsibilities
This software can access your Gmail inbox and Google Calendar through MCP connectors, and may save content from those sources as notes inside your local Obsidian vault.
You are solely responsible for:
a) Ensuring that your use of this software complies with all applicable privacy and data protection laws, including but not limited to:
- The EU General Data Protection Regulation (GDPR), Regulation (EU) 2016/679
- The California Consumer Privacy Act (CCPA), Cal. Civ. Code § 1798.100 et seq.
- The UK Data Protection Act 2018
- Any other national, regional, or sector-specific data protection legislation applicable to you
b) Obtaining all necessary consents and establishing a lawful basis before processing any personal data of third parties through this software.
c) Implementing appropriate technical and organizational measures to protect any personal data processed through this software, including but not limited to encryption, access controls, retention policies, and data subject rights mechanisms.
d) Responding to any data subject access requests, deletion requests, or other rights exercised by individuals whose data you process through this software.
4.2. Third-Party Data
Emails and calendar events contain personal data of other people (names, email addresses, phone numbers, meeting content, personal communications). When the Postman agent reads your inbox and saves content to your vault:
- That data is stored in plaintext on your local filesystem
- You become the data controller for that data under applicable law
- You are responsible for ensuring a lawful basis for that processing
- You are responsible for informing those individuals that their data is being processed, if required by law
- You are responsible for deleting that data upon request, if required by law
4.3. Author's Role
The author of this software:
- Is not a data controller or data processor with respect to any data you process
- Has no access to your vault, emails, calendar, or any data you process
- Does not collect, receive, store, or transmit any of your data or any third-party data
- Cannot respond to data subject requests on your behalf
- Accepts no liability for any data protection violations arising from your use of this software
5. AI-Generated Output and Model Limitations
5.1. Nature of the Output
This software provides prompts and instructions to large language models (LLMs). The author does not provide, host, train, fine-tune, or control any AI model. All output is generated by the underlying LLM at runtime.
You explicitly acknowledge and accept that:
a) Large language models are stochastic systems. They generate output probabilistically. The same input can produce different output on different runs. Output may contain errors, fabrications, contradictions, or nonsensical content at any time, without warning.
b) LLMs can and do hallucinate: they generate text that appears factual but is partially or entirely fabricated. This includes invented facts, false citations, incorrect calculations, non-existent references, and plausible-sounding but wrong information.
c) The author provides prompt engineering only. The behavior, accuracy, safety, and reliability of the output depend entirely on the underlying model, its training data, its alignment methods (RLHF, DPO, Constitutional AI, or others), its safety filters, and its runtime configuration. The author has no control over any of these factors.
d) This software is designed and tested with Anthropic's Claude models. If you use a different LLM (whether through a fork, a plugin modification, a different MCP configuration, or any other means), the quality, safety, and reliability of the output are entirely unpredictable and entirely your responsibility. Models without adequate alignment, safety training, or content filtering may produce harmful, misleading, dangerous, or offensive output.
e) Even when using Claude, no output should be treated as authoritative, factual, or reliable without independent verification. This applies to all agents — core and custom — including but not limited to: factual claims made by the Seeker, organizational suggestions by the Architect, and any advice-like output generated by custom agents in domains such as health, legal, financial, or any other regulated field.
f) Even though this software has been tested by the author using Anthropic's Claude, no quality standard can be guaranteed. Due to the inherent stochastic nature of large language models, the same prompt can produce different output across different sessions, model versions, context windows, and runtime conditions. Testing validates that the prompts are well-formed and produce reasonable results under observed conditions, but it cannot and does not guarantee consistent, accurate, or safe output in all circumstances or for all users.
5.2. No Responsibility for Model Behavior
The author:
- Does not guarantee the accuracy, completeness, safety, or appropriateness of any output generated by any agent under any circumstances
- Does not guarantee that safety instructions in the prompts will be followed by the model in all cases. Prompt-based safety is best-effort, not a guarantee
- Does not guarantee that the software will behave identically across different model versions, providers, or configurations
- Is not responsible for any output generated by models other than Anthropic's Claude, regardless of the reason the user chose to use a different model
- Is not responsible for any action you take based on AI-generated output
6. No Warranty
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT.
The author does not warrant that:
- The software will function correctly, without errors, or without interruption
- The output of any agent will be accurate, complete, or appropriate
- The software will not cause data loss or corruption in your vault
- The software is suitable for any particular purpose
7. Limitation of Liability
IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY CLAIM, DAMAGES, OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT, OR OTHERWISE, ARISING FROM, OUT OF, OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
This includes, without limitation, liability for:
- Any advice-like output generated by core or custom agents, including but not limited to health, legal, financial, dietary, and wellness domains
- Hallucinated, fabricated, inaccurate, or misleading content generated by the underlying AI model
- Any consequence of using this software with a model other than Anthropic's Claude
- Loss or corruption of data in your vault
- Violations of data protection law arising from your use of the software
- Any action taken by custom agents created by the user (Section 9)
- Any indirect, incidental, special, consequential, or punitive damages
- Loss of profits, data, goodwill, or other intangible losses
8. Forks, Modifications, and Derivative Works
The author takes no responsibility for forks, modifications, or derivative works of this software. If a third party modifies this software and distributes it, the modified version is governed by the third party's own terms, not these.
The author explicitly condemns any use of this software or its derivatives for: unauthorized data collection, psychological manipulation, impersonation of medical or mental health professionals, surveillance, or any form of exploitation or harm.
9. Custom Agents
This software allows users to create custom AI agents through the Architect agent. Custom agents are user-generated prompt configurations stored in the user's vault.
By creating and using custom agents, you explicitly acknowledge and accept that:
a) Custom agents are entirely your creation and your responsibility. The author of this software has no involvement in their design, behavior, or output.
b) The author provides no warranty whatsoever on the functionality, accuracy, safety, or reliability of any custom agent you create, regardless of whether the Architect agent assisted in generating the agent file.
c) You are solely responsible for everything your custom agents do, including but not limited to: what data they read, what notes they create, what files they modify, what information they process, and any consequences of their actions.
d) Custom agents are subject to the same limitations described in Section 5 (AI-Generated Output and Model Limitations). They produce stochastic, probabilistic output that may be inaccurate, incomplete, or misleading.
e) The author accepts no liability for any harm, data loss, privacy violation, or any other adverse outcome resulting from the use of custom agents, whether directly or indirectly.
f) If you share your custom agents with others (through forks, exports, or any other means), Section 8 (Forks, Modifications, and Derivative Works) applies in full. You become responsible for the shared agent's behavior.
10. Governing Law
These terms are governed by and construed in accordance with the laws of the jurisdiction in which the author resides, without regard to conflict of law principles. Any disputes arising from these terms or the use of the software shall be subject to the exclusive jurisdiction of the courts of that jurisdiction.
11. Changes to These Terms
The author reserves the right to modify these terms at any time. Changes will be reflected in the Last updated date at the top of this document. Continued use of the software after changes constitutes acceptance of the updated terms.
12. Acceptance
By using this software (including by running the installer, loading the plugin, or interacting with any agent), you confirm that you have read, understood, and agree to these Terms of Use.
If you create or use any custom agent that generates advice-like output in regulated domains (health, legal, financial, dietary, wellness, or similar), you additionally confirm that you have read and accepted the disclaimers in Section 3 above, and that you understand these disclaimers apply automatically to all such agents.