atlas-nav
Skill2 starsChanged 4 months agoMITView on GitHub
Use when navigating a repo's code or its Atlas — locating a definition/usage/file, tracing an import, or finding which zone/decision/spec explains something. Reach here before reflexive grep+Read.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- muslewski/token-oracleword for word, first seen Jul 28, 2026
Other versions1 project uses a version of this.
Word for word · 1
- muslewski/token-oracle★ 2100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- atlas-adoptSkill
- atlas-recollectionSkill
- atlas-seedSkill
- atlas-updateSkill
- writing-for-retrievalSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Default to this repo's configured retrieval engine over grep+read-whole-file — same correctness, far fewer tokens read.
- Very specific: line 47
If results look stale and this repo has the ctx-search adapter: node scripts/nav-refresh-index.mjs --force, then retry.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Navigating code and the Atlas
Default to this repo's configured retrieval engine over grep+read-whole-file
— same correctness, far fewer tokens read. If this repo uses the ctx-search
adapter (adapters/ctx-search/), that engine is ctx_search (the
context-mode MCP tool), kept fresh automatically by a SessionStart hook.
Otherwise use whatever retrieval engine the repo documents, or fall back to
grep — this skill's routing logic applies either way.
With ctx_search: call it with NO project param — it defaults to the
repo (cwd), which is where both buckets are indexed. Results are tagged
Source: code:… / Source: atlas:…; narrow with source: code /
source: atlas. Folders listed in atlas.config.json →
retrieval.excludeFromSearch (default drafts/, visuals/) are never in
the index — don't expect hits from them.
Route by intent
| Intent | Tool |
| Known file path | Read directly — no search |
| Exact symbol ("where is X defined / who calls X") | LSP / go-to-definition, if available |
| Broad ("where is X / which note explains Y / find the invariant / which spec") | Configured retrieval engine (ctx_search if this repo has the adapter) |
| Engine result empty/wrong, or an exhaustive match list is needed | grep + Read fallback |
Per-bucket mode
Atlas hits (Source: atlas:…) — the snippet usually captures the
zone/decision summary inline; no follow-up read needed for orientation.
Code hits (Source: code:…) — use the engine to RANK, then Read the top
~3 whole hit files; snippets alone miss the answer span.
Rules
- Query with distinctive content terms, not whole questions ("pending webhook retry queue", not "which note documents the pending webhook retry queue").
- Retrieval is lazy and per-question — reach for it freely, but NEVER preload the whole Atlas or code tree; that re-pays the tokens the engine just saved.
- If results look stale and this repo has the ctx-search adapter:
node scripts/nav-refresh-index.mjs --force, then retry.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

