Abstract-first retrieval
Scan the index before loading the book.
Agent memory, made inspectable
Turn ephemeral chats into reusable agent memory skills.
Long context is working memory. Skills are long-term memory. The Skill Memory Bank turns conversations into scoped, inspectable, reusable procedural memory that an agent can scan, pulse, and govern without flooding the context window.
Scan the index before loading the book.
Retrieve a path, not a pile of chunks.
Read, write, assess, correct, archive.
Abstracts are the index. Skills are the executable memory.
Ephemeral work, decisions, corrections, and outcomes.
Heuristics separate facts, preferences, procedures, and causes.
A portable, typed memory object with evidence and scope.
A tiny index card describing when the skill is useful.
Explicit entities activate a traversable neighborhood.
User, project, and sensitivity gates run before loading.
Use, correction, decay, merge, archive, and restore.
This deterministic local heuristic extracts entities, procedures, semantic preferences, causal hints, and a Level 5 abstract. No API, model, or server receives the text.
Use plan-first implementation, repo-local guidance, and verification loops before accepting agent-generated code.
Always inspect repo-local instructions before editing. · A green build is necessary but visual verification closes the loop.
Use plan-first implementation, repo-local guidance, and verification loops before accepting agent-generated code.
agent workflow, verification, repo guidance
Inspect repository guidance and conventions. · Write a concise implementation plan.
Use plan-first implementation, repo-local guidance, and verification loops before accepting agent-generated code.
skill://ContextJamming/codex-workflow/abstractSeeded fictional skills and browser-local additions share one searchable lakebed. Select a card to inspect its MCP resource and governance controls.
A graph pulse retrieves a path, not a pile of chunks. Entity anchors activate candidate skills, graph edges explain traversal, and scope gates run before any memory enters context.
Pulse an @mention to reveal the scored, scoped traversal bundle.
Memory must be scoped before it is smart. Graph pulses are intersected with user_id, project_id, and sensitivity constraints before context is assembled. The agent may discover that relevant memory exists without exposing the underlying content when scope is denied.
Keep discovery lightweight and execution explicit. Resources expose inspectable memory; tools perform lifecycle operations.
resources/listskill://{project}/{domain}/abstractskill://{project}/{domain}/tmt/L5skill://{project}/{domain}/proceduredistill_conversation_to_skillpulse_entity_networkassess_skill_utilitymerge_redundant_skillsarchive_low_utility_skill{
"uri": "skill://ContextJamming/codex-workflow/abstract",
"mimeType": "text/markdown",
"tokens": 28,
"scope": "demo-user:ContextJamming",
"utilityScore": 0.91
}Feedback changes what the system tries first. It does not rewrite reality. Human correction remains authoritative, and every memory needs a visible exit.
No assessments recorded for this skill yet.
Illustrative, conservative arithmetic—not a benchmark. The model compares replaying every prior transcript with scanning abstracts and loading a few complete skills.
Instead of replaying 12 full sessions, the agent scans 18 compact indexes and recursively loads 3 complete skills. Real savings depend on transcript, tokenizer, and retrieval behavior.
A useful memory substrate preserves different kinds of knowledge without pretending they are interchangeable.
Events, outcomes, exceptions, and sequence.
Preferences, facts, principles, and constraints.
Repeatable steps, checks, and decision rules.
Utility, correction, decay, merge, and archive.
| Dimension | Full-context memory | Flat vector RAG | Skill Memory Bank |
|---|---|---|---|
| Token cost | Grows with every session | Moderate; repeated chunks | Abstract scan + selective load |
| Causal traceability | Buried in transcript | Usually weak | Explicit causal edges |
| Procedural reuse | Implicit | Fragmented | Executable workflow |
| Scope isolation | Prompt-dependent | Filter-dependent | Gate before retrieval |
| Human correction | Appended | Re-embedded | Lifecycle action |
| Multi-session compounding | Context bloat | Index growth | Skill refinement |
A production-intent artifact needs an adversarial review loop. This portable brief asks another agent to inspect the live behavior, infer the implementation, prioritize risks, and recommend the next high-leverage iteration.
You are reviewing a production-intent POC for ContextJamming called **Skill Memory Bank**. **Live page:** https://www.contextjamming.com/SkillMemoryBank **Project context (for you):** - This is a browser-local simulation (no server, no external APIs) that turns ephemeral agent conversations into scoped, inspectable, reusable procedural memory "skills". - Core principles: abstract-first retrieval, @mention graph pulses, user-governed lifecycle, and strict scoping before intelligence. - It must feel like a real memory substrate, not just marketing copy. - Visual/typographic language should stay consistent with Context Jamming (editorial, high-signal, Fraunces + IBM Plex Mono influence, clean cards, clear hierarchy). - This POC is meant to be both a compelling demo for investors/partners and a working primitive we can evolve toward real agent use (MCP resources/tools, future sovereign stacks). **Your task — thorough code + UX review:** 1. **Analyze the live page thoroughly** (all sections, all interactive elements, buttons, cards, the distillation example, graph pulse area, safety boundary table, governance feedback, import/export/clear). 2. **Infer or request the implementation details** you need (HTML structure, Tailwind usage, vanilla JS / state model, localStorage schema, distillation heuristic logic, graph representation, how pulsing and scoping actually work in code, any seeded data vs dynamic creation). 3. **Deliver a structured review** with the following sections: **A. Executive Summary** (2-3 sentences on overall quality, fidelity to the 7-stage vision, and production readiness) **B. Strengths** (What is already working well — architecture, UX moments, conceptual clarity, code patterns worth keeping) **C. Issues & Risks** (categorized + prioritized) - P0 (must fix for this to be a credible demo) - P1 (important for maintainability / extensibility) - P2 (nice-to-have polish) Categories to cover: - Conceptual fidelity (does the implementation actually demonstrate abstract-first + graph pulsing + scoped governance, or is it mostly static explanation?) - Distillation heuristic quality & transparency - Graph pulse / retrieval implementation - Data model & localStorage design (schema, versioning, migration path) - State management & reactivity - Code organization & maintainability (especially if still monolithic single-file) - UX / interactivity gaps (empty states, feedback, real skill creation flow, verification loops) - Accessibility & keyboard support - Mobile / responsive behavior - Error handling, edge cases, and "what if" scenarios - Performance / DOM bloat risks - Security / scoping simulation robustness **D. Specific Recommendations** For each major issue, give concrete suggestions (and code snippets where helpful). Prioritize changes that increase the "this feels like real governed memory" perception. **E. Quick Wins** A short list of high-impact, low-effort improvements that would make the POC feel significantly more alive. **F. Extensibility Notes** How easy/hard would it be to: - Add real user-created skills from pasted conversation - Evolve the graph into a proper traversable structure - Wire this to a real MCP server later - Reuse components/patterns in future ContextJamming POCs **G. Final Verdict + Recommended Next Step** (One paragraph + a clear suggested scope for the next iteration) **Tone & Approach:** - Be direct but constructive. You are helping ship a high-signal artifact. - Reference the "Codex Workflow Hardening" skill principles where relevant (plan-first, repo-local guidance, verification loops, visual + build verification). - Assume we will iterate quickly — focus on leverage, not perfectionism. Begin your review now. If you need the full HTML source or specific sections of the JS, tell me exactly what to paste.
This POC is a conversation artifact: a way to show how personal and organizational memory can become inspectable, portable, governed agent skill infrastructure.
§ · Invoice No. 001 · The Build Ledger
Filed · contextjamming.com
What a conservative mid-market digital agency would have quoted for the same scope, itemized against what this site actually cost. Agency numbers are the floor — not the premium brand-studio tier.
TIME
12 weeks
2 days
~42× faster
COST
~$150,000
~$300
~500× cheaper
TEAM
5-person agency
1 human + 3 models
Same deliverable
§ Itemized — what a mid-market agency SOW would have billed
Agency figure assumes ~700 billable hours at $200/hr blended, plus ~18% PM overhead — the conservative floor of a mid-market SOW. Premium brand studios would have quoted 2–3× that. Stack: Antigravity (orchestrator), Claude Opus 4.8 (auditor), Codex (adversary), Cloudflare Workers / OpenNext.
§ Colophon
Vol. 26 · build log
Every page on contextjamming.com is the output of a real-time, three-body Mixture-of-Experts loop. One model orchestrates. Two consult. The human holds the thesis. No single model commits alone.
View Redesign Assessment →Orchestrator
Google DeepMind
Auditor
1M context
Adversary
Cross-model MoE
Stack
Typeset in
Infrastructure
human intent
│
▼
┌────────────────────┐ ┌─────────────────┐
│ Antigravity │ ◄────► │ Claude Opus 4.8 │ ← auditor loop
│ (orchestrator) │ │ (auditor) │
└─────────┬──────────┘ └─────────────────┘
│ ◄───────────┐
▼ │
┌──────────┐ ┌────┴───────┐
│Cloudflare│ │ Codex │ ← adversarial loop
│ Workers │ │ │
└─────┬────┘ └────────────┘
│
▼
contextjamming.com
│
▼
┌──────────────┐
│ Git push │ ← audit trail
└──────────────┘