SKF Knowledge Overview

Principle

Agent-level knowledge files capture cross-cutting principles and patterns that Ferris applies across multiple workflows. They are distinct from workflow references/ and assets/ files, which contain step-specific operational artifacts. Knowledge is loaded just-in-time when a workflow step directs — never preloaded into context.

Rationale

The SKF module has 14 workflows containing operational references/ and assets/ files within each skill directory. References contain rules, patterns, and protocols for decision-making; assets contain templates, schemas, and output formats. Both are tightly coupled to individual steps. But several principles — zero hallucination, confidence tiers, provenance — cut across multiple workflows and need a single authoritative source rather than fragmented repetition.

Without a knowledge folder:

• Cross-cutting principles get duplicated across workflow steps
• Updates to foundational rules require changes in multiple locations
• Ferris has no centralized reference for judgment calls that span workflows

With knowledge files:

• Each principle has one canonical definition
• Workflow steps reference knowledge files instead of re-stating rules
• Ferris loads only the knowledge relevant to the current task

Knowledge Map

File	Purpose	Key Workflows
zero-hallucination.md	Every claim traces to source; uncitable content excluded	AN, BS, CS, QS, SS, US, AS, TS, RA
confidence-tiers.md	T1/T1-low/T2/T3 trust model and citation formats	AN, BS, CS, QS, SS, US, AS, TS, EX, RA, VS
progressive-capability.md	Quick/Forge/Forge+/Deep tier philosophy and behavior adaptation	SF, AN, BS, CS, QS, SS, US, AS, TS, EX, VS, RA
agentskills-spec.md	agentskills.io output format principles and compliance	CS, QS, SS, US, TS, EX
skill-lifecycle.md	End-to-end pipeline — workflow connections, artifact flow, decision rules for workflow selection, pipeline invocation syntax and aliases	All 14
provenance-tracking.md	Provenance-map.json entries and file_entries, evidence reports, claim traceability	CS, QS, SS, US, AS, TS
manual-section-integrity.md	[MANUAL] marker preservation and merge algorithm	CS, US, AS, TS, EX
qmd-registry.md	Progressive QMD collection registry and collection gate principle	SF, BS, CS, US, AS
ccc-bridge.md	ccc_bridge interface, indexing lifecycle, Forge+ discovery-before-extraction	SF, AN, CS, SS, US, AS
tool-resolution.md	Bridge-to-tool and subprocess-to-tool mapping per IDE environment	CS, QS, SS, US, AS, AN, BS, RA
architecture-verification.md	Pre-code verification pipeline, integration verdicts, iteration loop	VS, RA, SS
doc-fetcher.md	T3 external documentation fetching, quarantine rules, tool-agnostic	CS, QS
split-body-strategy.md	Selective vs full split-body, detection pattern, agent accuracy	CS, QS, US, TS, EX
version-paths.md	Version-aware storage layout, path templates, export manifest v2, migration	CS, QS, SS, US, AS, TS, EX, VS, RA
overview.md	Knowledge map and JiT loading protocol (this file)	All (entry point)

JiT Loading Protocol

1. Ferris receives a task or workflow activation
2. Consult skf-knowledge-index.csv to identify relevant knowledge fragments by tags or description. The tier column classifies fragments as core (always relevant when the domain applies) or extended (load only when the specific scenario is active — e.g., doc-fetcher.md only when doc_urls are present in the brief)
3. Load only the files needed for the current task — never bulk-load all knowledge
4. Apply knowledge principles during workflow execution alongside step-specific references/ and assets/ files

Knowledge vs. Data

Aspect	Knowledge Files (knowledge/)	References (references/)	Assets (assets/)
Scope	Cross-cutting, multi-workflow	Single workflow, single step	Single workflow, single step
Content	Principles, patterns, judgment rules	Rules, patterns, protocols, heuristics	Templates, schemas, output formats
Loading	JiT via index lookup	JiT via step instruction	JiT via step instruction
Examples	Zero hallucination, confidence tiers	scoring-rules.md, extraction-patterns.md	skill-brief-schema.md, skill-template.md
Stability	Rarely changes	Changes with workflow evolution	Changes with workflow evolution

Related Fragments

• confidence-tiers.md — the trust model that underpins extraction and scoring
• skill-lifecycle.md — the end-to-end pipeline connecting all workflows
• progressive-capability.md — tier philosophy that affects every workflow

Source: synthesized from module.yaml, README.md, and workflow README