media-mind-forge
Analyze, introspect, and distill foundational learnings from video transcripts, lectures, and media into executable agent skills and Knowledge Items (KIs). Combines book-to-skill-forge procedural synthesis with mind-reader epistemic introspection. Use when analyzing video transcripts, distilling mental models from lectures, extracting Knowledge Items with isnad lineage, or forging new skills from recorded media.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- media-mind-forge
- description
- Analyze, introspect, and distill foundational learnings from video transcripts, lectures, and media into executable agent skills and Knowledge Items (KIs). Combines book-to-skill-forge procedural synthesis with mind-reader epistemic introspection. Use when analyzing video transcripts, distilling mental models from lectures, extracting Knowledge Items with isnad lineage, or forging new skills from recorded media.
02 资源文件 14
- 脚本__init__.py1000 B
- 参考文档CARD.md5.0 KB
- 脚本engine/__init__.py1.2 KB
- 脚本engine/brief.py5.7 KB
- 脚本engine/core.py7.4 KB
- 脚本engine/distillation.py8.5 KB
- 脚本engine/models.py1.3 KB
- 脚本engine/procedures.py6.6 KB
- 脚本engine/scaffolder.py6.5 KB
- 脚本engine/transcripts.py4.0 KB
- 脚本main.py11.6 KB
- 脚本models.py7.8 KB
- 参考文档plugin.json4.7 KB
- 参考文档README.md5.6 KB
03 SKILL.md 正文
Media Mind Forge: Cognitive Introspection & Skill Synthesis
media-mind-forge is the authoritative multi-modal learning engine for Brain Harness. It unifies the procedural methodology extraction of book-to-skill-forge with the epistemic introspection and isnad provenance tracking of mind-reader to transform spoken media, recorded lectures, and video transcripts into durable Knowledge Items (KIs) and executable agent skills (SKILL.md + CARD.md).
Every media forging session adheres to three foundational pillars:
- The Visual Brief — Interactive HTML reports generated in
%TEMP%featuring Mermaid cognitive topology diagrams. - The Mandatory Checkpoint — Explicit human-in-the-loop gates (
RequestFeedback: true) before committing synthesized skills or mutating knowledge stores. - Explicit Anti-Patterns — Rigid behavioral boundaries eliminating superficial summaries, speculative over-commitment, and ungrounded claims.
See CARD.md for the companion summary card, stage matrix, and invariants checklist.
Semantic Intent Descriptors & Routing Triggers
The Agent Skill Knowledge Graph routes tasks to this skill when any of the following triggers or conceptual intents are detected:
- Primary Triggers:
"analyze video transcript","learn from video","distill lecture","forge skill from transcript","media mind forge" - Downstream Operations:
"extract mental models from video","create knowledge items from lecture","transcribe and learn","synthesize coaching rubric from video" - Topological Successor To:
youtube-transcript-fetcher,stagehand_browser,web_fetcher - Topological Predecessor To:
skill_knowledge_graph,hermes_state_fts5,epistemic-memory-lifecycle
The Dual-Lens Cognitive Extraction Model
┌─────────────────────────────────────────────────────────────────────────┐
│ MEDIA MIND FORGE DUAL-LENS ENGINE │
├────────────────────────────────────┬────────────────────────────────────┤
│ Lens 1: Mind-Reader Introspection │ Lens 2: Procedural Skill Forge │
│ - First Principles & Mental Models │ - Shu-Ha-Ri Execution Stages │
│ - Decision Heuristics & Trade-offs │ - Diagnostic Coaching Questions │
│ - Epistemic Isnad Lineage Blocks │ - Anti-Pattern Defense Matrices │
│ - Knowledge Items (KIs) to Vault │ - Executable SKILL.md & CARD.md │
└────────────────────────────────────┴────────────────────────────────────┘
Execution Sequence
Stage 1: Ingest & Deconstruct Transcript
Ingest raw transcript text, timestamped segment arrays, or fetch live via youtube-transcript-fetcher. Normalize formatting, compute speaking pace and total duration, and isolate core procedural sections from conversational filler.
Completion criterion: Verified transcript text and timestamped segments parsed with word count and duration recorded.
Stage 2: Dual-Lens Extraction & Epistemic Isnad Binding
Execute simultaneous Mind-Reader and Forge analysis:
- Epistemic Distillation (Mind-Reader): Unpack foundational axioms, decision heuristics, and delta learnings. Bind every claim to exact transcript timestamps (
isnad_claims). - Procedural Synthesis (Forge): Formulate 3 to 6 operational stages with deterministic completion gates and author diagnostic interview scorecards.
Completion criterion: Mental models, heuristics, procedural stages, diagnostic rubric, and Knowledge Items formulated.
Stage 3: The Visual Synthesis Brief
Generate an interactive, self-contained HTML report at %TEMP%/media-mind-forge-<timestamp>.html rendering a Mermaid Cognitive Topology DAG connecting first principles to decision trade-offs and operational stages.
graph TD
Transcript[Spoken Media Transcript] --> MindReader[Lens 1: Epistemic Introspection]
Transcript --> Forge[Lens 2: Procedural Synthesis]
MindReader --> KIs[Knowledge Vault: Ground-Truth KIs]
Forge --> Skill[Executable Agent Skill: SKILL.md + CARD.md]
KIs --> Router[Harness Skill Knowledge Graph]
Skill --> Router
Completion criterion: Self-contained HTML report generated in
%TEMP%and presented with clickable link.
Stage 4: Mandatory Checkpoint Gate
Author an implementation_plan.md artifact with RequestFeedback: true detailing proposed Knowledge Items, skill name, and target directory. STOP and await explicit user approval before writing files to persistent stores.
Completion criterion: Explicit user confirmation received before committing artifacts.
Stage 5: Artifact Scaffolding & Knowledge Vault Commit
Write the validated SKILL.md and CARD.md into the target skill directory (enforcing single-pipe │ borders and standard anti-patterns heading per Rule 37). Persist extracted Knowledge Items to storage and register in the Skill Knowledge Graph.
Completion criterion: Validated skill package authored and Knowledge Items indexed with 100% test pass rate.
Anti-Patterns
- Superficial Narrative Summarization — Paraphrasing transcript dialogue into passive summaries without extracting operational algorithms or mental models. Always extract checkable Shu-Ha-Ri execution stages and first-principle heuristics.
- Ungrounded Isnad Claims — Asserting axioms or authoritative advice without citing supporting transcript quotes or video timestamps. Always bind claims to timestamped isnad nodes.
- Speculative Tool Sprawl — Creating fragmented single-purpose scripts for each video instead of modular, reusable skill architectures. Deepen existing skills in-place or generate standardized deep-module packages.
04 同仓库更多 Skills
ui-behavior-primitive-architect
Architect, audit, and verify zero-dependency headless UI behavior primitives, WebMCP agent bridges, WAI-ARIA/SSR/RTL tri-modal checks, and framework expert routing. Do not use for static CSS theme styling or backend database migrations.
trustgraph-ontology-governor
Govern knowledge graphs, SHACL-AF triple policies, OntologyRAG subclass validation, PROV-O dual named graphs, and Laya entropy decision scoring. Do not use for relational SQL queries or generic web scraping.
shim-harness-governor
Govern universal CLI agent loops, 4-tier tool permissions, ancestor-symlink workspace sandboxes, deterministic-first command routing, collision-safe MCP namespacing, and layered skill loading. Do not use for static UI styling or external SQL analytics.
graphiti-zep-memory-governor
Govern bi-temporal knowledge graphs with Graphiti future-edge clamping, MinHash/LSH entity dedup, RRF/MMR reranking, and Zep typed ontologies. Do not use for static SQL joins or generic web crawling.
agora-mcp-workbench-architect
Architect, audit, and orchestrate Agora Workbench MCP servers, loopback sidecars, unified state graphs, gateway/dispatcher connectors, and policy-enforced data lakes. Do not use for static frontend CSS styling or non-MCP desktop GUI automation.
skill-eval-assurance-runner
Evaluate and assure agent skills using small-N Wilson confidence intervals, unbiased pass^k reliability, deterministic trajectory graders, preflight boundary honesty, and sealed digest verification. Do not use for subjective prose review or unconstrained production deployment.