Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Searches and navigates stored knowledge in memory palaces. Use when looking for previously stored information or cross-referencing concepts across palaces.
.claude/skills/athola-knowledge-locator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 84% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 98% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 109% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 66% | 0% |
A spatial indexing and retrieval system for finding information within and across memory palaces. Enables multi-modal search using spatial, semantic, sensory, and associative queries.
The Knowledge Locator provides efficient information retrieval across your memory palace network by:
bashpython scripts/palace_manager.py search "authentication" --type semantic
Verification: Run python --version to verify Python environment.
bashpython scripts/palace_manager.py list
Verification: Run python --version to verify Python environment.
palace structures - use memory-palace-architect
external resources - use knowledge-intake
palace structures - use memory-palace-architect
external resources - use knowledge-intake
| Mode | Description | Best For | |------|-------------|----------| | Spatial | Query by location path | "Find concepts in the Workshop" | | Semantic | Search by meaning/keywords | "Find authentication-related items" | | Sensory | Locate by sensory attributes | "Blue-colored concepts" | | Associative | Follow connection chains | "Related to OAuth" | | Temporal | Find by creation/access date | "Recently accessed" |
modules/index-structure.mdmodules/search-strategies.mdmodules/index-structure.mdSearch the review chamber within project palaces for past decisions and patterns.
bash# Search review chamber by query python scripts/palace_manager.py search "authentication" \ --palace <project_id> \ --room review-chamber # List entries in specific room python scripts/palace_manager.py list-reviews \ --palace <project_id> \ --room decisions # Find by tags python scripts/palace_manager.py search-reviews \ --tags security,api \ --since 2025-01-01
Verification: Run python --version to verify Python environment.
| Room | Content | Example Query | |------|---------|---------------| | decisions/ | Architectural choices | "JWT vs sessions" | | patterns/ | Recurring solutions | "error handling pattern" | | standards/ | Quality conventions | "API error format" | | lessons/ | Post-mortems | "outage learnings" |
When starting work in a code area, surface relevant review knowledge:
bash# When in auth/ directory python scripts/palace_manager.py context-search auth/ # Returns: # - Past decisions about authentication # - Known patterns in this area # - Relevant standards to follow
Verification: Run python --version to verify Python environment.
Works with:
memory-palace-architect - Indexes palaces created by architectsession-palace-builder - Searches session-specific palacesdigital-garden-cultivator - Finds garden content and linksreview-chamber - Searches PR review knowledge in project palacesscripts/palace_manager.py search "<query>" returns resultswithin 500ms for a cold query against an indexed palace
sensory, associative, temporal) returns ranked results for a valid query against an existing palace
the queried concept is present in the indexed palace
satisfying the ≥ 80% robustness target
memory-palace-architect rather than returning an empty result silently
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | fail→pass | 6,795 | 2,348 | -65% | 1 | 1 | 0% | 963 | 1,771 | +84% | 0 | 0 | — |
case-19 | fail→pass | 5,262 | 1,482 | -72% | 1 | 1 | 0% | 799 | 1,581 | +98% | 0 | 0 | — |
case-20 | pass→pass | 7,662 | 1,707 | -78% | 1 | 1 | 0% | 1,251 | 1,549 | +24% | 0 | 0 | — |
case-01 | fail→pass | 11,532 | 6,208 | -46% | 1 | 1 | 0% | 1,981 | 2,582 | +30% | 0 | 0 | — |
case-02 | fail→pass | 7,447 | 5,464 | -27% | 1 | 1 | 0% | 1,125 | 2,356 | +109% | 0 | 0 | — |
case-03 | fail→pass | 22,122 | 8,469 | -62% | 1 | 1 | 0% | 1,723 | 2,862 | +66% | 0 | 0 | — |
case-04 | fail→pass | 9,557 | 3,021 | -68% | 1 | 1 | 0% | 1,461 | 1,936 | +33% | 0 | 0 | — |
case-05 | fail→pass | 13,077 | 2,865 | -78% | 1 | 1 | 0% | 2,079 | 1,803 | -13% | 0 | 0 | — |
case-06 | fail→pass | 11,545 | 3,854 | -67% | 1 | 1 | 0% | 2,106 | 2,039 | -3% | 0 | 0 | — |
case-07 | fail→pass | 8,961 | 2,304 | -74% | 1 | 1 | 0% | 1,490 | 1,805 | +21% | 0 | 0 | — |
case-08 | pass→pass | 26,235 | 1,736 | -93% | 1 | 1 | 0% | 2,289 | 1,648 | -28% | 0 | 0 | — |
case-09 | pass→pass | 8,889 | 2,008 | -77% | 1 | 1 | 0% | 1,394 | 1,669 | +20% | 0 | 0 | — |
case-10 | pass→pass | 9,763 | 2,239 | -77% | 1 | 1 | 0% | 1,445 | 1,723 | +19% | 0 | 0 | — |
case-11 | pass→pass | 9,615 | 2,210 | -77% | 1 | 1 | 0% | 1,542 | 1,686 | +9% | 0 | 0 | — |
case-12 | fail→pass | 8,505 | 3,253 | -62% | 1 | 1 | 0% | 1,312 | 1,952 | +49% | 0 | 0 | — |
case-13 | pass→pass | 8,137 | 2,088 | -74% | 1 | 1 | 0% | 1,274 | 1,601 | +26% | 0 | 0 | — |
case-14 | pass→pass | 8,935 | 1,966 | -78% | 1 | 1 | 0% | 1,452 | 1,671 | +15% | 0 | 0 | — |
case-15 | pass→pass | 9,012 | 2,057 | -77% | 1 | 1 | 0% | 1,394 | 1,675 | +20% | 0 | 0 | — |
case-16 | fail→pass | 8,629 | 1,599 | -81% | 1 | 1 | 0% | 1,457 | 1,625 | +12% | 0 | 0 | — |
case-17 | pass→pass | 7,812 | 3,253 | -58% | 1 | 1 | 0% | 1,387 | 1,826 | +32% | 0 | 0 | — |
case-21 | pass→pass | 6,515 | 2,177 | -67% | 1 | 1 | 0% | 1,159 | 1,688 | +46% | 0 | 0 | — |
case-22 | fail→pass | 8,748 | 2,117 | -76% | 1 | 1 | 0% | 1,455 | 1,735 | +19% | 0 | 0 | — |
case-23 | pass→pass | 7,666 | 1,944 | -75% | 1 | 1 | 0% | 1,156 | 1,684 | +46% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +52 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.