菜单
Use the infer skill to generate speculative inference candidates from claims, persist them on branch-local inference graphs, and run the full mutation pipeline that coordinates active memory, claims, Terminus persistence, and speculative ranking.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use the speculate skill when you need critique-ready hypotheses from new observations: recall context, run the mutation pipeline on a fresh observation, retrieve ranked inference and facet results from an inference/* branch, and present candidate text, provenance, assumptions, uncertainty, and review priority without promoting anything to trusted memory.
Use the discover skill to find connections between claims through facet relations and to rank inference and facet candidates using manifold geometry. Covers facet modeling, manifold scoring, relatedness, distance, uncertainty, and retrieval of discovery results.
Use the memorize skill to store observations, entities, tasks, and claims into the filesystem-backed active memory. Covers session lifecycle, working-set writes, entity/task card creation, claim extraction, and candidate promotion.
Use the recall skill to retrieve composed context from active memory, trusted Terminus graphs, and optional speculative branches. Also covers journal-backed history of past working-memory mutations.
Use the reflect skill to persist memories, claims, and inference nodes into the TerminusDB temporal graph. Covers branch lifecycle, trusted session writes, temporal recall, and the separation between session and inference branches.
| name | infer |
| description | Use the infer skill to generate speculative inference candidates from claims, persist them on branch-local inference graphs, and run the full mutation pipeline that coordinates active memory, claims, Terminus persistence, and speculative ranking. |
Generate speculative knowledge from existing claims and persist it on isolated inference branches.
import pathlib
from src.active_memory.models import WorkingItem
from src.persistence.pipeline import MutationPipeline
from src.terminus.adapter import TerminusMemoryRepository
from src.manifold_sidecar import ManifoldRankingService
root = pathlib.Path(".agent-memory")
repo = TerminusMemoryRepository(url="http://localhost:6363")
ranker = ManifoldRankingService()
pipeline = MutationPipeline(
root,
enable_terminus=True,
terminus_repo=repo,
manifold_service=ranker,
enable_inference=True,
)
item = WorkingItem(
item_type="observation",
content="The auth service failed after a certificate rotation.",
session_id="sess-1",
)
result = pipeline.run(item, session_id="sess-1")
print(result.inference_candidates) # number of candidates generated
print(result.ranked_inference_candidates) # number ranked by manifold sidecar
print(result.inference_branch) # e.g. "inference/sess-1"
from src.inference.generator import InferenceGenerator
from src.claims.extractor import ClaimExtractor
extractor = ClaimExtractor()
claims = extractor.extract(text="The API started returning 401 errors.")
generator = InferenceGenerator()
candidates = generator.generate_from_claims(
claims,
source_branch="inference/sess-1",
source_commit="journal-event-abc",
session_id="sess-1",
)
facets = generator.generate_facet_candidates(
claims,
provenance_commit="journal-event-abc",
source_branch="inference/sess-1",
)
| Class | Method | Purpose |
|---|---|---|
MutationPipeline | run() | Full pipeline: active write → journal → claims → Terminus → inference → ranking |
InferenceGenerator | generate_from_claims() | Produce InferenceNode candidates from a list of claims |
InferenceGenerator | generate_facet_candidates() | Produce FacetRelation candidates between adjacent claims |
src/persistence/pipeline.py — mutation pipeline orchestrationsrc/inference/generator.py — rule-based inference and facet generationsrc/inference/models.py — InferenceNode and FacetRelation schemassrc/claims/extractor.py — upstream claim extractiontests/integration/test_inference_flow.py — integration coveragenotebooks/02_speculative_inference_and_facets.ipynb — executable walkthroughMutationPipeline in src/persistence/pipeline.py so all pipeline steps stay synchronized.InferenceGenerator from src/inference/generator.py for candidate creation.inference/* branches; never write directly to trusted memory.tests/integration/test_inference_flow.py.