fidelity-ai-workspace/agent-memory/integrations/obsidian.md

type, audience, tags

type	audience	tags
guide	workspace-user	onboarding obsidian

Obsidian Adapter

Use Obsidian to navigate and review project knowledge. Do not use it as a separate memory database.

Obsidian is the current interface over canonical Markdown memory. The workspace-level abstraction is scripts/memory/, so agents should use Obsidian-specific commands only through that adapter when possible.

---

Recommended Start

Open the workspaces/fidelity/project-knowledge/ folder as the Obsidian vault, not the repository root and not agent-memory/.

Open:

• Start Here
• New Member Onboarding
• Knowledge Maps

---

Best Views

New Member View

Start with:

• New Member Onboarding
• Glossary
• Current Work Map
• Fidelity Domain Map

Daily Work View

Start with:

• Current Work Map
• Work Items Map
• Daily Notes Index

System Understanding View

Start with:

• Fidelity Domain Map
• Fidelity Apps Map
• People Map

---

Graph Guidance

Use the graph to find clusters, not to infer truth.

High-value graph hubs:

• start-here
• onboarding
• current-work
• fidelity-domain
• work-items
• people
• fidelity-apps
• fidelity-domain

If a node is isolated, it may still be valid. It may be:

• a daily log
• raw evidence
• a command/prompt file
• a file that needs to be linked from a map

---

Global Graph Defaults

The global graph is configured for onboarding and project navigation, not full filesystem inspection.

It intentionally hides generic technical filenames such as:

• index
• README
• AGENTS
• SKILL
• 08-bases
• 09-templates

Those files still exist and remain valid workspace files. They are hidden from the graph because Obsidian displays only the basename, which creates many indistinguishable nodes.

Use the file explorer or search when you need a specific technical file. Use the graph when you want to understand relationships between project concepts.

---

When To Show Technical Files

Temporarily remove these filters from the graph search when debugging workspace internals:

-file:index -file:README -file:AGENTS -file:SKILL

For normal onboarding, keep them hidden and navigate through the named maps.

---

Editing Rules

• Edit canonical memory files when the fact is clear.
• Prefer updating existing notes over adding duplicate summaries.
• Do not edit generated files as durable memory.
• Do not treat inbox files as confirmed truth.
• Use maps to improve navigation, not to store detailed facts.

---

Daily Notes, Templates, And Bases

• Daily notes live under 06-daily/ and use 09-templates/daily.md.
• Templates live under 09-templates/.
• Bases live under 08-bases/ and use simple type properties plus a folder exclusion for 09-templates/.
• Templates may contain the final note type so newly created notes are useful immediately, but Bases must exclude the template folder so template files do not appear as data rows.
• Role mapping files such as 04-people/manager.md should use type: role-map, not type: person.
• Bookmarks should keep 00-start/start-here.md, 00-start/onboarding.md, current work, work items, people, and high-value Bases easy to open.

---

CLI And Templates

The official Obsidian CLI can create notes from templates, query Bases, inspect properties, and search the vault.

In this workspace, prefer the platform-agnostic memory wrapper:

bash scripts/memory/memory.sh create work-item pdiap-15999 "Example title"
bash scripts/memory/memory.sh search "PDIAP-15765"
bash scripts/memory/memory.sh base-query work-items
bash scripts/memory/memory.sh health

The wrapper owns note type routing, so a work-item goes to 02-work-items/, a person goes to 04-people/, a decision goes to 05-decisions/, and so on.

Use Obsidian CLI directly only when debugging Obsidian-specific behavior:

bash scripts/obsidian/cli.sh help