agentleFS
Sign inStart free
Tyler-Hughes312Tyler-Hughes312/mycelium/CLAUDE.md

mycelium

CLAUDE.md1 starsChanged 3 months agoMITView on GitHub

This repo runs a local Mycelium Core (http://127.0.0.1:8787) that provides RAG-scoped, token-budgeted context — the same chunk/embed/retrieve/budget-assemble pipeline used by services/core/src/mycelium/core/domain/chatassembler.py and ragservice.py, but applied to the coding session itself instead of the chat product. Use it instead of broad grepping or dumping whole files. - At the start of meaningful work, call myceliumsessionstart (or myceliumpreflight for lighter exploration) with the absolute workspace path. It returns a compact bootstrap packet plus a one-line receipt — cite the receipt, don't re-fetch…

We split this CLAUDE.md into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are 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.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before planning or building anything new, call mycelium_reuse_check(goal) first — it checks for prior art across all indexed repos.

    line 14
  2. Very specific:

    Prefer mycelium_change_context(goal) when implementing, mycelium_debug_context(error) when fixing, and mycelium_search(query) for general …

    line 16
  3. Distinctive:

    Do not dump chat transcripts into the vault, and do not paste full files when a receipt already covers the hit list.

    line 18
  4. Very specific:

    If mycelium_health reports the Core is unreachable, fall back to normal exploration (Explore agent / grep) rather than blocking on it.

    line 21

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 CLAUDE.md 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

MemoryOptimization

Mycelium context layer (required)

This repo runs a local Mycelium Core (http://127.0.0.1:8787) that provides RAG-scoped, token-budgeted context — the same chunk/embed/retrieve/budget-assemble pipeline used by services/core/src/mycelium/core/domain/chat_assembler.py and rag_service.py, but applied to the coding session itself instead of the chat product. Use it instead of broad grepping or dumping whole files.

  • At the start of meaningful work, call mycelium_session_start (or mycelium_preflight for lighter exploration) with the absolute workspace path. It returns a compact bootstrap packet plus a one-line receipt — cite the receipt, don't re-fetch the full vault/code.
  • Before planning or building anything new, call mycelium_reuse_check(goal) first — it checks for prior art across all indexed repos. If it says to ask the user reuse-vs-new, ask and wait.
  • Prefer mycelium_change_context(goal) when implementing, mycelium_debug_context(error) when fixing, and mycelium_search(query) for general lookups — over grepping or globbing many files.
  • Do not dump chat transcripts into the vault, and do not paste full files when a receipt already covers the hit list.

If mycelium_health reports the Core is unreachable, fall back to normal exploration (Explore agent / grep) rather than blocking on it.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

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.

Check your setup, free