메뉴
Brain knowledge base operations. The core read/write cycle: brain-first lookup, read-enrich-write loop, source attribution, ambient enrichment, back-linking. Read this before any brain interaction.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/praxstack/skills-and-personas --skill brain-ops명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Create exhaustive BLUEPRINT.md files — the implementation bible that expands a SPEC.md into maximum granular detail so an AI coding agent can translate it line-by-line into code with zero interpretation. This skill REQUIRES a SPEC.md to already exist. Use this skill whenever the user wants to create a blueprint, implementation blueprint, detailed spec, "bible" document, or says "create the blueprint", "expand the spec", "make it more detailed", "generate BLUEPRINT.md", or "I need every detail spelled out." Also trigger when the user has a SPEC.md and wants to go deeper before implementation. The Blueprint is the final artifact before code — it takes the portable implementation contract (SPEC.md) and fills in every edge case, validation rule, error message, sequence, and example so nothing is left to judgment. Works for any software domain. If no SPEC.md exists, this skill MUST redirect the user to generate one first using the spec-creator skill.
Create production-grade SPEC.md files — portable implementation contracts that any AI coding agent can build from without follow-up questions. Use this skill whenever the user wants to write a spec, specification, technical spec, implementation contract, SPEC.md, or service specification for any software task. Also trigger when the user says "spec out", "write a spec for", "create a spec", "define the contract for", or describes a system they want built and needs a thorough specification before implementation. This skill works for backend services, full-stack apps, CLI tools, SDKs, developer tools, and any domain. It produces specs at the level of a principal-engineer-authored portable implementation contract — not a PRD, not an HLD, but the artifact that sits between architecture and code.
Scope-calibrated ultra-rigor workflow for hard reasoning, high-risk code changes, architecture decisions, debugging with multiple plausible root causes, security-sensitive work, and user requests like "ultra reasoning", "think harder", "verify everything", "adversarial review", "war room", "deep check", "paranoid verifier", or "no hallucinations". Use to force evidence-first planning, assumption tracking, hypothesis falsification, adversarial self-review, verification gates, and clear uncertainty without overloading trivial tasks.
SKILL.md 표시 중
| name | brain-ops |
| version | 1.1.0 |
| upstream | brain-ops@fc834ee |
| description | Brain knowledge base operations. The core read/write cycle: brain-first lookup, read-enrich-write loop, source attribution, ambient enrichment, back-linking. Read this before any brain interaction. |
| triggers | ["any brain read/write/lookup/citation"] |
| tools | ["search","query","get_page","put_page","add_link","add_timeline_entry","get_backlinks","sync_brain"] |
| mutating | true |
| writes_pages | true |
| writes_to | ["people/","companies/","deals/","concepts/","meetings/"] |
The brain is not an archive. It is a live context membrane that every interaction flows through in both directions.
Convention: See
skills/conventions/brain-first.mdfor the 5-step lookup protocol. Convention: Seeskills/conventions/quality.mdfor citation and back-link rules.
Memory verbs (MEMORY_VERBS v1, gbrain ≥ 0.43). Over MCP, prefer the five frozen memory verbs for the read/write cycle:
remember(fact, provenance, ttl?)to save a single durable fact (mandatory provenance; dedupes + supersedes),recall(query | entity, budget_tokens)to read it back budget-packed,entity(name)for a zero-LLM card,synthesize(question)for the expensive cross-page answer,forget(id)to expire a fact. Userememberinstead ofextract_factswhen you already have ONE formed fact;put_page/add_link/add_timeline_entrystay the page/graph write path. Fall back to the classic ops when the verbs aren't on the surface. Contract:docs/protocol/MEMORY_VERBS_v1.md.Keyless brains: when
extract_factsreturnsskipped: extraction_unavailable, YOU are the extractor — pull the facts from the turn yourself and write each one viarememberwithkindset (event | preference | commitment | belief) and the visibility the envelope'sagent_actionnames (default private — pin it;rememberdefaults to world), or author a## Factsfence on the entity page. Askipped: extraction_failedenvelope (server-side extractor errored on this turn;reasonnames why) invites the same manualrememberfallback for that turn — automatic extraction stays on for future writes.
This skill guarantees:
[Source: ...] citations)Every mention of a person or company with a brain page MUST create a back-link
FROM that entity's page TO the page mentioning them. An unlinked mention is a
broken brain. See skills/conventions/quality.md for format.
Before using ANY external API to research a person, company, or topic:
gbrain entity "<name>" (v0.43+) — ONE known person/company/project → full card (description, aliases, open threads, recent events, edges, backlink/fact counts). Zero LLM calls, sub-100ms. This one call replaces steps 2–6 for known-entity lookups; near-misses return suggestions.gbrain search "name" — exact-token lookup for existing pages (cheap hybrid, no expansion)gbrain query "natural question about name" — concept/landscape questions go here FIRST (expansion recovers synonym phrasings; a nonzero search count is not proof of completeness)gbrain get <slug> — if you know the slug, read the full pageThe brain almost always has something. External APIs fill gaps, not start from scratch.
⚠️ NEVER scope/count a corpus with shallow ls — query gbrain or find. Federated sources often carry MULTIPLE coexisting directory conventions — a flat legacy layer AND a date-nested meetings/YYYY/MM/ layer. A non-recursive ls dir/*.md sees only one and undercounts massively. Real example: a shallow ls of one source's meetings/ counted 132 files, almost all the user's, and concluded that WAS the corpus — missing thousands of transcripts nested under meetings/YYYY/MM/. To count/scope a brain corpus:
gbrain sources list (shows per-source indexed page counts) + gbrain query. gbrain indexes ALL federated sources correctly; trust its index, not the filesystem.find <dir> -name '*.md' | wc -l, never ls *.md. Then map the layout: find <dir> -name '*.md' | sed -E 's#(.*/)[^/]+$#\1#' | sort | uniq -c.gbrain sources list before believing a low count.For questions that need synthesis, temporal grounding, or analytical answers — not just "find the page" but "answer the question":
gbrain think "<question>" — multi-hop synthesis across pages + takes +
the graph. Temporal questions route through trajectory analysis; everything
else gets an LLM-synthesized, cited answer with conflict + gap analysis.
Returns a grounded answer, not just a list of matching pages.query for
simple page lookups where you just need the slug or a quick context check.Every message, meeting, email, or conversation that references a person or company:
User's direct statements are the highest-value data source. Write them to brain
pages immediately with attribution [Source: User, YYYY-MM-DD].
Every put_page call automatically extracts entity references and writes them
to the graph (links table) with inferred relationship types. Stale links
(refs no longer in the page text) are removed in the same call. This is
"auto-link" reconciliation.
add_link calls needed for ordinary page writes.attended (meeting -> person), works_at, invested_in,
founded, advises, source (frontmatter), mentions (default).put_page MCP response includes auto_links: { created, removed, errors }
so the agent can verify outcomes.gbrain config set auto_link false. Default is on.gbrain timeline-add
(or batch via gbrain extract timeline --source db).Before answering any question about a person, company, or topic:
Don't answer from general knowledge when a brain page exists.
This is not a special mode. This is the default. Everything the user says is an ingest event.
Rules:
No separate output. Brain-ops is an always-on behavior layer, not a report generator. The output is updated brain pages and enriched responses.
When a brain has multiple sources (wiki, gstack, yc-media, etc.), every
citation MUST include the source id: [source-id:slug]. Example:
You told me about the retry budget approach — see [wiki:topics/resilience] and [gstack:plans/retry-policy] for where this came from.
Rules:
sources.id (immutable), never sources.name (mutable display).[default:slug] OR may omit the prefix
for backward compat.search, query, get_page, list_pages
carries source_id — always use it when citing, never guess.If a search result has source_id: "gstack" and slug: "plans/foo",
the citation is [gstack:plans/foo]. That's the whole rule.
[Source: ...] citationsgbrain entity "<name>" (catches aliases + near-misses), then query with name variantssearch — cheap hybrid search (vector + keyword, no expansion)query — hybrid search + LLM multi-query expansion (concept/landscape questions)get_page — read a brain pageput_page — create/update brain pagesadd_link — cross-reference entitiesadd_timeline_entry — record eventsget_backlinks — check who references an entitysync_brain — sync changes to the index