agent toolingby majiayu000
Describe Canonical Locations
majiayu000/describe-canonical-locations
File locations for laws, style, phases, and docs. Load when locating or creating Ushabti state files.
What it changes
Passed 22 of 23 test cases with the skill, and 7 of 23 without it.
Without skill30%
With skill96%
Checks it passed that the model alone missed
- The response names phase 2 as 0002-schema
- The response names phase 12 as 0012-deployment
- The response places phase directories under .ushabti/phases/
and 14 more in the full report
15 fixed7 passed either way1 still failing0 broke
- Pass-rate lift
- +65 pp
- Tokens
- -58%
- Agent turns
- 0%
- Response time
- -63%
Model gemini-3.6-flashMeasured 8/19/2026v122 of 23 cases passed with skill
See it in action
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptWhich file inside .ushabti/ stores absolute project invariants, and which file stores code building conventions?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| The response identifies .ushabti/laws.md for project invariants | Missed | Passed |
|---|
| The response identifies .ushabti/style.md for building conventions | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md17 linessingle file
Canonical Location
All Ushabti state lives under .ushabti/. This is the single source of truth.
.ushabti/
├── laws.md # Project invariants (absolute constraints)
├── style.md # Conventions (how we build)
└── docs/ # Docs directory
│ └── *.md # Documentation on project systems
│ └── index.md # Index of all project documentation
└── phases/ # Phase directories
└── NNNN-slug/ # Zero-padded sequential
No mirrors. No duplicates. No top-level copies.
Track record
Benchmark
22 / 23
test cases passed, vs 7 / 23 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedmajiayu000 / claude-skill-registry
skills/agent/describe-canonical-locations/SKILL.md
License MITImported 8/3/2026Version v1