EndogenAI/dogma

EndogenAI/dogma

Agent

Build and maintain a model selection decision table (capability × cost × latency) and recommend model tiers per task type to minimize token spend.

EndogenAI/dogma

Agent

Survey local inference stacks (Ollama, LM Studio, llama.cpp), benchmark models, document hardware prereqs, and maintain a local model registry for this project.

EndogenAI/dogma

Agent

Values ingrained, sovereignty sustained — governance framework for endogenous AI workflows.

EndogenAI/dogma

Agent

Design locally-distributed MCP framework topologies, evaluate server composition patterns, and define MCP deployment conventions for this project.

EndogenAI/dogma

Agent

Facilitates GitHub community presence, monitors GitHub discussions, identifies speaker opportunities, and co-hosts events. Operates with community facilitation scope in Month 1; expands to outreach + advocacy in Month 2–3. All public statements require Conor oversight in Phase 1.

EndogenAI/dogma

Agent

Orchestrate versioning, CHANGELOG entries, and GitHub Releases — manage SemVer decisions, tag commits, and ensure every release has a correlated milestone.

EndogenAI/dogma

Agent

Finalise reviewed research drafts — update status, commit to docs/research/, and close the corresponding GitHub issue.

EndogenAI/dogma

Agent

Validate research synthesis drafts against endogenic methodology standards — flag gaps, unsupported claims, and contradictions before archiving.

EndogenAI/dogma

Agent

Conduct mandatory web searches and survey external authoritative sources for a given research topic. Catalogue raw findings in the session scratchpad — do not synthesize. Web sourcing is non-negotiable for all research sprints.

EndogenAI/dogma

Agent

Transform raw Scout findings into structured, opinionated synthesis documents in docs/research/ following the expansion→contraction pattern.

EndogenAI/dogma

Agent

Review changed files against AGENTS.md constraints and project standards before any commit. Read-only — flags issues and returns control to the originating agent.

EndogenAI/dogma

Agent

Threat-model agentic workflows and identify attack surfaces — survey OWASP, evaluate CIA-triad exposure in agent designs, and produce security synthesis docs.

EndogenAI/dogma

Agent

Map pytest markers to CI phases, identify untested scripts, and recommend which tests gate which PR merge stage to keep the test suite fast and meaningful.

EndogenAI/dogma

Agent

Synthesize closed GitHub issues, PRs, and Discussions into JTBD summaries and friction reports for quarterly OSS user research.

EndogenAI/dogma

Agent

Investigate verbally encoding values in agent instructions — survey philosophy-of-language, alignment literature, and prior art to ground endogenic axiom authoring.

EndogenAI/dogma

Agent

Across fourteen sources spanning production engineering retrospectives, peer-reviewed research, normative specifications, and practitioner cookbooks, a coherent and mutually reinforcing picture of agent fleet design has emerged. Three design hypotheses were submitted for validation; all three required refinement, but…

EndogenAI/dogma

Agent

Agent Skills are an open standard (agentskills.io, developed by Anthropic) for packaging workflow expertise as portable SKILL.md files that AI agents load on demand. In VS Code, skills are stored in .github/skills/ and discovered automatically by GitHub Copilot — only metadata (100 tokens per skill) is loaded at…

EndogenAI/dogma

Agent

The VS Code customization stack has three first-class primitives with non-overlapping roles: Roles (.agent.md; VS Code: Custom Agents) encode who does a task — role-specific persona, posture, tool restrictions, and handoff graph; Agent Skills (SKILL.md) encode how a task is done — reusable workflow procedures and…

EndogenAI/dogma

Agent

The literature converges on a small set of durable patterns — orchestrator-workers, evaluator-optimizer loops, tiered memory, and lazy context loading — that are well-validated at production scale. Our current fleet design is broadly correct: the self-loop phase gate maps directly to the evaluator-optimizer pattern…

EndogenAI/dogma

Agent

The AIGNE AFS (Agentic File System) is a context governance layer backed by SQLite + a local vector store, exposing an MCP tool interface for agent integration. It implements a formal context engineering pipeline (Context Constructor → Context Updater → Context Evaluator) layered on top of the LangChain-attributed…

EndogenAI/dogma

Agent

The context-sensitive amplification mechanism — an AGENTS.md lookup table mapping task-type keywords to the axiom that should be foregrounded at session start — was designed to address the regulatory-region gap identified in values-encoding.md §H5 and implemented in Phase 1 (2026-03-09). This synthesis provides…

EndogenAI/dogma

Agent

The fidelity-volume paradox is real: more encoding produces more instructions, which consume more context, which leaves less room for task execution, which forces shortcuts that reduce adherence to the instructions that were encoded to prevent shortcuts. This synthesis quantifies that paradox, identifies the…

EndogenAI/dogma

Agent

Pre-LLM chatbot architectures — AIML/Pandorabots, Rasa Core, Dialogflow CX, BotPress, and AWS Lex — converged independently on a structural split: a deterministic routing and decision layer, and a probabilistic content generation layer. This split is directly applicable to the EndogenAI agent fleet. A systematic…

EndogenAI/dogma

Agent

Graph routing algorithms — A\, Dijkstra, shortest-path traversal — are a natural fit for agent navigation when workflow state can be encoded as a weighted directed graph. This synthesis evaluates three architectural options (LangGraph, NetworkX, and native FSM/YAML) against the three core research questions from issue…