Menu
Load Parzival cross-session memory from Qdrant
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Detect content drift of an operator's scaffolded sanctum files (BOND, CAPABILITIES, CREED, INDEX, LORE, MEMORY, PERSONA, PULSE) against the evolving reference templates, and surface recommended add/remove WITH rationale — never a silent overwrite. Use on a session-start drift check, after the reference templates change, or when the operator asks whether their sanctum is current.
Check ai-memory system status and collection stats
Check ai-memory system status and collection stats
Manually save current session context to ai-memory
Search ai-memory for relevant stored memories
Search ai-memory for relevant stored memories
| name | aim-parzival-bootstrap |
| description | Load Parzival cross-session memory from Qdrant |
| allowed-tools | Bash |
This skill is invoked only on demand from [ST] Session Start workflow step-01b. It does NOT auto-load at activation.
parzival.md activation step 4 verifies the presence of the core skills (aim-parzival-bootstrap, aim-parzival-constraints, aim-parzival-loader) in {skills_path} without reading their content. Per parzival.md <rules> r6 ("Load files ONLY when executing user-chosen workflow"), skill content is read only when the menu command that uses the skill is invoked.
Load cross-session context from previous Parzival sessions stored in Qdrant. This replaces the automatic startup injection with an on-demand skill invocation.
The script retrieves three layers from Qdrant and emits each as a markdown section:
### Last Handoff — most recent agent handoff (deterministic, agent_id=parzival)### Recent Decisions — recent decisions and agent memories (deterministic)### Insights — recent agent insights (deterministic, agent_id=parzival)Sanctum Tier B files (LORE.md, BOND.md) are not loaded here — they load in step-01b section 4 (Tier B), after the Qdrant gate completes.
"${AI_MEMORY_INSTALL_DIR:-$HOME/.ai-memory}/.venv/bin/python" \
"${AI_MEMORY_INSTALL_DIR:-$HOME/.ai-memory}/_ai-memory/pov/skills/aim-parzival-bootstrap/bootstrap.py"
Include the script output in your current context as cross-session memory from previous Parzival sessions.
If the script reports Qdrant unavailable or an error, note this and continue with file-based context only (MEMORY.md, oversight/ files).
_ai-memory/pov/skills/aim-parzival-bootstrap/bootstrap.pyresolve_project_id (env-first → cwd/git → fail-loud), the same helper the write paths use, so read and write never disagree (BUG-314)