agentleFS
DioNanosDioNanos/nexuscrew/skills/vl-msa/SKILL.md

vl-msa

Skill4 starsChanged 36 days agoApache-2.0View on GitHub

Use when indexing or retrieving documents, notes, research or past conversations through a VL-MSA long-term-memory MCP server, including collections, batch indexing, BM25 or hybrid search, full-document grounding, remember/forget and bounded multi-hop interleaving. Enforce locate-then-ground retrieval and choose the user-facing language from the request.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 6 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use VL-MSA for durable searchable source material.

    line 8
  2. Distinctive:

    Keep collection IDs, document IDs, tool names, metadata fields and quoted source text unchanged unless the user explicitly asks to …

    line 20
  3. Very specific:

    Call msa_search {collection, query, k} to locate relevant chunks.

    line 27
  4. Very specific:

    Call msa_fetch_doc for the selected hit before using it as evidence.

    line 28
  5. Distinctive:

    Search chunks are locators, not complete sources.

    line 30
  6. Very specific:

    Use msa_index for one document or note.

    line 35
  7. Very specific:

    Use msa_index_batch for bulk ingest.

    line 36
  8. Very specific:

    Use msa_remember for a standalone agent memory.

    line 46
  9. Very specific:

    Use msa_forget for an explicitly requested removal.

    line 48
  10. Distinctive:

    Do not claim a memory was stored or removed until the tool confirms the outcome.

    line 50
  11. Very specific:

    Use msa_interleave_round for one bounded route, deduplicate and read step.

    line 55
  12. Distinctive:

    Use BM25 by default.

    line 61
  13. Distinctive:

    Do not describe lexical-only results as semantic retrieval.

    line 64
  14. Very specific:

    If no VL-MSA tool is available and this skill is packaged with NexusCrew, the optional companion is documented in ../../MCP_COMPANIONS.md.

    line 66

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

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

VL MSA

Use VL-MSA for durable searchable source material. Collections persist across sessions; BM25 works without embeddings, while hybrid reranking is optional.

Select response language

Choose the language for explanations and answers in this order:

  1. the user's explicit language preference;
  2. the language of the current request;
  3. a reliable client or system locale;
  4. English.

Keep collection IDs, document IDs, tool names, metadata fields and quoted source text unchanged unless the user explicitly asks to translate them.

Locate, then ground

Retrieval has two required steps:

  1. Call msa_search {collection, query, k} to locate relevant chunks.
  2. Call msa_fetch_doc for the selected hit before using it as evidence.

Search chunks are locators, not complete sources. Do not answer a context-dependent question from truncated snippets alone.

Index material

  • Use msa_index for one document or note.
  • Use msa_index_batch for bulk ingest.
  • Group related material into a stable named collection.
  • Inspect msa_list_collections, msa_stats or msa_manifest before assuming a collection exists or is populated.

Index only material the user is authorized to store. Preserve provenance and do not place credentials or unrelated personal data in collection metadata.

Manage standalone memories

  • Use msa_remember for a standalone agent memory. Its deduplication and low-signal gate may decline a write.
  • Use msa_forget for an explicitly requested removal.

Do not claim a memory was stored or removed until the tool confirms the outcome.

Handle multi-hop questions

Use msa_interleave_round for one bounded route, deduplicate and read step. Repeat only as needed, carrying forward the grounded facts from the prior round. Avoid a single unbounded search.

Choose ranking

  • Use BM25 by default.
  • Pass dense_alpha only when the server is built and configured for hybrid retrieval.
  • Do not describe lexical-only results as semantic retrieval.

If no VL-MSA tool is available and this skill is packaged with NexusCrew, the optional companion is documented in ../../MCP_COMPANIONS.md. Explain the missing capability and ask before installing or configuring anything.

Dependencies

Bundled: nothing — this skill is documentation only.

External (you must provide):

NeedInstallProbe / failure mode
A VL-MSA MCP server (companion)optional companion mcp-vl-msa-rs: https://github.com/DioNanos/mcp-vl-msa-rs — the repository's Install section is authoritative and may change; this skill deliberately does not duplicate its commands (not verified here)if the companion is not registered, the client exposes no msa_* tools; ask the client for its tool list
Client MCP registrationregister the companion's stdio command in the AI client's MCP configsame failure mode: absent tools, not a runtime error

Without the companion, durable retrieval is simply unavailable in that session: say so and suggest the companion once instead of reading the server's on-disk collections directly (which this skill forbids).

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.

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.