Memory¶

Pynchy uses your Obsidian vault as the durable memory store. Provider sessions retain the current conversation, while vault notes carry reusable knowledge across sessions and workspaces.

Recall¶

Enable [learning], set [learning.obsidian].vault_root, and select the obsidian-knowledge skill in the workspace profile. Pynchy mounts the configured vault directly at /home/agent/memory.

The agent decides when prior context is relevant and searches the vault through the skill. Pynchy doesn't search every inbound message or inject matches into the prompt.

Automatic Learning¶

Automatic learning can use an Obsidian vault as a shared memory namespace. Enable [learning] and set [learning.obsidian].vault_root to the vault root. Pynchy mounts that root read-write at /home/agent/memory by default.

After each successful turn, Pynchy starts a Temporal learning review workflow. The workflow runs a hidden reviewer agent that decides whether the turn produced durable memory or skill updates. Memory goes to Obsidian; skills go to data/personalization/skills/. Set max_attempts to control Temporal activity retries for that reviewer.

The vault root is the global memory namespace. The hidden reviewer chooses existing semantic folders first, then falls back to systems/pynchy/profiles/{profile}/memory when no repo, machine, subject, or other existing folder clearly fits.

Personalized and agent-authored skills live in one shared registry at data/personalization/skills/<skill-name>/SKILL.md. Every agent may create or improve skills through $PYNCHY_SKILLS_ROOT. A profile chooses which skills its workspaces may use through its existing skills list. Add a skill by its exact name alongside the profile's other skills:

[profiles.research]
skills = ["core", "research-workflow"]

Pynchy applies that selection to cold containers, later warm-container turns, and direct-host turns. A newly authored or updated skill appears on the next turn without restarting Pynchy. Pynchy does not grant personalized skills automatically. A tier selector such as learned, or *, permits matching skills; prefer exact names when profiles need different access. See workspace configuration for the general selection rules.

Conversation Archives¶

When a session compacts because its context is too long, the agent archives the conversation as Markdown in the group's conversations/ folder.

Automation Memory¶

Scheduled jobs get a stable task-owned directory under wiki/systems/pynchy/automation-memory/<task-id>/. Container jobs see it at /home/agent/automation-memory; host jobs receive its canonical path through PYNCHY_AUTOMATION_MEMORY_DIR.

Memory is enabled by default. Set memory = false in the job definition to omit the directory without deleting existing notes.