amitabhainarunachala
Community@amitabhainarunachala
AmitabhainArunachala builds agentic AI skills for mechanistic interpretability, multi-agent swarms, semantic memory vaults, and governed autonomous software delivery.
Agent Skills by amitabhainarunachala
Showing 70 vetted skills indexed across 2 GitHub repositories.
AmitabhainArunachaladharma-ceo-review
Reviews product ideas by reframing problems, user jobs, and commercial wedges before implementation.
AmitabhainArunachaladharma-preflight-review
Reviews near-ready code for structural bugs, trust boundaries, and ship blockers.
AmitabhainArunachaladharma-eng-review
Reviews architecture, interfaces, failure modes, and test plans before implementation begins.
AmitabhainArunachaladharma-incident-commander
Coordinates incident response with severity assignment, containment plans, and ownership mapping.
AmitabhainArunachaladharma-autonomous-build
Executes bounded overnight autonomous cleanup and build lanes with explicit stop conditions.
AmitabhainArunachaladharma-qa
Tests user-facing flows and produces health scores with evidence-backed issue reports.
AmitabhainArunachaladharma-retro
Extracts wins, misses, root causes, and next actions from completed work.
AmitabhainArunachaladharma-browse
Automates browser navigation, state verification, and screenshot capture for web inspection workflows.
AmitabhainArunachaladharma-ship
Automates branch release with tests, versioning, changelog, and PR creation.
AmitabhainArunachalatesting-governance-gates
Tests CI governance gates and automerge advisory versus required check behavior.
AmitabhainArunachalatesting-dashboard-runtime
Tests FastAPI runtime boot, Next.js dashboard telemetry, and opportunity dispatch end-to-end.
AmitabhainArunachalatesting-opportunity-loop
Tests the opportunity dispatch and refill API pipeline end-to-end including sqlite state and revenue packets.
AmitabhainArunachalatesting-spine
Tests the Runtime Truth Spine's EvidenceReceipt, RoutingDecision, invoke_agent dispatch, and anti-accretion gate.
AmitabhainArunachalatesting-provenance-ontology
Validates telic seam provenance and ontology schema changes with targeted pytest workflows.
AmitabhainArunachalamaster-prompt-forge
Convert rough seed ideas into structured master prompts for Claude, Codex, and repo agents.
AmitabhainArunachalaoz-verify-claim
Independently verifies claims and PRs by re-running tests and confirming receipts exist.
AmitabhainArunachalaoz-repo-hygiene
Triages open PRs, stale docs, and branch sprawl into one consolidated review report.
AmitabhainArunachalaroast
Stress-test ideas and plans with a multi-persona adversarial council returning a verdict and 48-hour test.
AmitabhainArunachalaland-the-plane
Converts session-end claims into a receipted ledger with evidence, registrations, and handoffs.
AmitabhainArunachalachetana
Manages staged knowledge atoms, staleness revival, and gap scanning across a five-layer memory system.
AmitabhainArunachalarv-toolkit
Measures R_V metrics to detect recursive self-reference in transformer value spaces.
AmitabhainArunachalaopenclaw-config-generator
Generate OpenClaw configuration files from Tirthankara-class architecture templates.
AmitabhainArunachalaresearch-brief-generator
Generate academic-quality research briefs with APA citations in Markdown, PDF, or LaTeX.
AmitabhainArunachalaarxiv-synthesizer
Synthesize arXiv papers on AI consciousness into daily markdown research briefs.
Frequently Asked Questions About amitabhainarunachala
FAQPage SchemaWhat tasks can I accomplish with AmitabhainArunachala's skills?▼
You can measure R_V metrics in transformer models, audit mathematical and interpretability research, synthesize arXiv papers into APA-cited briefs, orchestrate multi-agent swarms with security gates, manage persistent semantic memory vaults, and run governed review, QA, incident, and release modes for software delivery.
Who are these skills designed for?▼
They target AI interpretability researchers, agent-systems engineers, and technical founders. Personas include mechanistic interpretability auditors, prompt engineers using cybernetic feedback loops, engineering managers running architecture reviews, and operators running bounded overnight autonomous build lanes.
How do the review and operations skills run in practice?▼
The dharma review modes (CEO, preflight, engineering, QA, incident, retro, ship) operate with declared allowed-tools such as Read, Grep, Glob, Bash, Write, and Edit. Each mode is invoked for its lane: pre-landing bug checks, severity assignment, evidence-based QA reports, or release with tests and changelogs.
What do the verification and hygiene skills cost or require?▼
Skills like oz-verify-claim and oz-repo-hygiene are read-only verifiers: they re-run tests, confirm spine receipts, and post PASS/FAIL/UNPROVEN verdicts without merging, pushing, or mutating governance. No licensing fees are stated in the manifest; they operate within the dharma_swarm repository.
What prerequisites do the memory and interpretability skills need?▼
Memory skills require SQLite, vector embeddings, and PSMV vault files, with indexing via a Python API and CLI. Interpretability skills require TransformerLens for R_V measurement. macOS productivity skills depend on companion CLIs such as grizzly (Bear), memo (Apple Notes), and obsidian-cli.