Skip to content

zy839971925-zyy/Agent--deep-research-Workflow-Skills / integrations/deepseek-harness/skills/reasoning-workflow

reasoning-workflow

Verified

Default silent workflow for everyday questions, research, writing, design and engineering. Match effort to the request, resolve costly ambiguity, find specialist resources and verify the requested outcome. Explicitly selectable with “调用我的 workflow skills”.

FileKindSize
RELEASE.jsonconfig509 B
SKILL.mdskill md5.4 KB
UPGRADE.mdmarkdown13.1 KB
assets/icon.svgother3.1 KB
families/audit-verification.mdmarkdown1.1 KB
families/core-reasoning.mdmarkdown2.0 KB
families/decision-analysis.mdmarkdown698 B
families/deep-research.mdmarkdown2.2 KB
families/execution-control.mdmarkdown1.5 KB
families/workflow-learning.mdmarkdown500 B
references/change-governance-and-effectiveness.mdmarkdown12.0 KB
references/clarification-gate.mdmarkdown6.9 KB
references/comparative-synthesis.mdmarkdown3.2 KB
references/decision-and-recommendation.mdmarkdown6.6 KB
references/domain-patterns.mdmarkdown4.2 KB
references/engineering-delivery.mdmarkdown2.8 KB
references/evidence-and-provenance.mdmarkdown7.3 KB
references/formal-review-and-audit.mdmarkdown7.4 KB
references/harness-integration.mdmarkdown7.5 KB
references/harness-profiles.mdmarkdown11.8 KB
references/host-adapter.mdmarkdown11.8 KB
references/human-context-and-interpretation.mdmarkdown3.9 KB
references/hypotheses-and-bias-control.mdmarkdown6.2 KB
references/information-target-navigation.mdmarkdown7.4 KB
references/inquiry-and-research.mdmarkdown9.1 KB
references/measurement-and-operationalization.mdmarkdown3.1 KB
references/model-adaptation.mdmarkdown7.5 KB
references/multimodal-and-data.mdmarkdown3.4 KB
references/problem-framing-and-effort.mdmarkdown8.6 KB
references/reasoning-structure-and-decomposition.mdmarkdown5.5 KB
references/relationships-and-systems.mdmarkdown4.2 KB
references/research-modes-and-methods.mdmarkdown17.1 KB
references/research-to-report.mdmarkdown3.8 KB
references/retrieval-and-observation.mdmarkdown14.4 KB
references/runtime-and-delegation.mdmarkdown8.9 KB
references/semantic-state-contract.mdmarkdown5.7 KB
references/specialist-resources.mdmarkdown2.8 KB
references/state-model-and-invariants.mdmarkdown7.1 KB
references/synchronization-and-recovery.mdmarkdown9.6 KB
references/synthesis-execution-and-verification.mdmarkdown6.7 KB
references/task-strategies.mdmarkdown4.9 KB
references/time-scenarios-and-forecasting.mdmarkdown3.6 KB
references/traceability-and-integrity.mdmarkdown6.2 KB
references/ultra-research.mdmarkdown9.1 KB
references/worked-examples.mdmarkdown5.1 KB
references/workflow-learning-and-capa.mdmarkdown4.4 KB
references/writing-for-the-reader.mdmarkdown1.7 KB
routing-index.jsonconfig18.0 KB
schemas/capability-snapshot.schema.jsonconfig1.1 KB
schemas/checkpoint-manifest.schema.jsonconfig5.6 KB
schemas/delivery-manifest.schema.jsonconfig4.9 KB
schemas/dual-verification.schema.jsonconfig2.6 KB
schemas/edge-policy.jsonconfig3.8 KB
schemas/event.schema.jsonconfig1.9 KB
schemas/execution-node-state.schema.jsonconfig2.7 KB
schemas/execution-plan.schema.jsonconfig5.7 KB
schemas/execution-schedule.schema.jsonconfig3.7 KB
schemas/learning-record.schema.jsonconfig4.5 KB
schemas/raw-record.schema.jsonconfig3.3 KB
schemas/review-bundle.schema.jsonconfig1.6 KB
schemas/task-profile.schema.jsonconfig4.2 KB
schemas/work-state.schema.jsonconfig90.9 KB
schemas/worker-proposal.schema.jsonconfig3.1 KB
scripts/build_distribution.pyscript1.7 KB
scripts/check_report_evidence.pyscript3.6 KB
scripts/harness_bridge.pyscript9.2 KB
scripts/host_adapter.pyscript2.1 KB
scripts/learning_core.pyscript6.3 KB
scripts/prepare_task.pyscript2.6 KB
scripts/reference_core.pyscript1.4 KB
scripts/requirements.txtmanifest37 B
scripts/retrieve_learnings.pyscript1.6 KB
scripts/route_resources.pyscript4.1 KB
scripts/route_task.pyscript1.0 KB
scripts/routing_core.pyscript6.5 KB
scripts/runtime_core.pyscript36.4 KB
scripts/semantic_core.pyscript14.9 KB
scripts/validate_delivery.pyexecutablescript10.4 KB
scripts/validate_dual_verification.pyscript3.1 KB
scripts/validate_events.pyscript2.8 KB
scripts/validate_learning.pyscript718 B
scripts/validate_raw.pyscript3.0 KB
scripts/validate_runtime.pyexecutablescript2.5 KB
scripts/validate_state.pyexecutablescript3.3 KB
scripts/validate_workflow.pyscript9.0 KB

references/workflow-learning-and-capa.md

View on GitHub
> Upgrade 2.1: read the updated [host contracts](host-adapter.md) and [migration rules](../UPGRADE.md) for executable admission, review and storage semantics. The shorter root SKILL is the active entry. Older examples are descriptive and must satisfy the current schemas.

# Workflow Learning and CAPA

[← Back to Workflow Learning family](../families/workflow-learning.md) · [Back to root workflow](../SKILL.md)

Use this module only in the maintenance plane, after a run or evaluation has produced external outcome evidence, deterministic findings, independent review, user correction, benchmark results, or another source of genuinely new information.

## Core boundary

Self-learning is not self-editing. A model criticizing its own answer without new evidence may create a **candidate insight**, but cannot validate or promote a canonical learning.

Treat reusable rules and Skills as **earned, not merely written**: a useful-looking heuristic from one run remains a candidate until outcome evidence, counterexamples, regression comparison, and applicability boundaries justify promotion.

## CAPA failure loop

For material workflow nonconformity use:

`failure → containment → root cause → corrective action → regression → verification → effectiveness check → preventive/generalization → closure`

Classify root cause before changing instructions. Useful classes include problem misframing, wrong depth, wrong routing, context overload, missing context, evidence/provenance failure, instruction conflict, model capability limit, tool/runtime failure, validator gap, execution-plan error, schedule/delegation error, and checkpoint/recovery error.

Do not respond to every failure by adding prompt text. Correct the mechanism that allowed the failure.

## Learning tiers

### Tier 1 — Experience memory

Store only validated, portable strategy/recovery/optimization lessons. Retrieve a small number whose applicability conditions match the active Task Profile and current gap. Track whether use helped or harmed.

### Tier 2 — Routing/reference heuristic

A routing lesson may propose changes to depth or reference selection. Promotion requires held-out evaluation against the current baseline and negative controls.

### Tier 3 — Canonical workflow change

Changes to SKILL.md, schemas, policy, validators, closure, authorization or runtime invariants require:

`candidate → minimal regression fixture → offline/shadow eval → baseline comparison → independent review → promote/reject → rollback available`

Never promote a Tier 3 change from a single trajectory.

## Memory hygiene

Validated memory may be promoted, downgraded, deprecated or marked stale. Deprecated/stale lessons are not runtime-retrievable. Prefer compact transferable heuristics over transcript storage. Preserve provenance and counterexamples.

## Applicability challenge and negative transfer

Retrieval is not permission to apply a lesson. Before using a remembered strategy, compare the lesson's applicability conditions and anti-conditions with the current Task Profile, environment, model/runtime capabilities, objective, and failure mode.

Use the smallest decision:

`use | adapt | ignore`

Prefer ignoring a superficially similar lesson when the differences affect its mechanism. Track harmful reuse as negative-transfer evidence and downgrade or narrow the lesson rather than adding compensating prompt text around it.

## Learn from three outcome shapes

Reusable lessons can come from:

- **clean success:** strategy or routing lesson;
- **failure → recovery → success:** recovery lesson;
- **successful but wasteful trajectory:** optimization lesson.

Do not reward a strategy merely because the final task succeeded if it consumed unnecessary context, calls, retries, or review. Efficiency lessons still require regression/quality gates so optimization cannot silently reduce answer quality.

## Learning and CAPA integration

CAPA creates a learning candidate only after corrective action passes verification and an effectiveness observation supports generalization. A one-off fix is not automatically a reusable lesson.

## Runtime isolation

Learning Plane is not part of normal task context. Ordinary runs may retrieve only a small validated experience-memory subset selected from the active profile and gap. Canonical Skill edits remain maintenance operations.

---

**Routing rule:** Return to the governing router after this module. `Related ≠ automatic load`; do not recursively load references.
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall