zedarvates
- 83 skills
- 0 followers
- 2 repo stars
- 6 days ago last updated
- ▌ Universal Compressor · zedarvates bundleCompacter des données choisies pour le contexte : JSON, logs et sorties d'outils. Restauration optionnelle en mémoire ; conserver les instructions et le code exacts.
- ▌ Vector Protocol · zedarvates bundleEncode structured findings into 24 numeric positions and search them in the in-memory VectorStore. This experimental codec uses fixed features, not learned embeddings; decoding cannot recover original paths or descriptions. Preserve the original findings and measure results before integration.
- ▌ Token Compressor · zedarvates bundleEncode UTF-8 text with the reversible TC1 zlib/base85 codec and decode it with TokenCompressor.expand. All accepted level names use this same codec. Use for encoded transport with a compatible decoder, not direct model-readable context compaction or a promise of fewer model tokens.
- ▌ Context Budget · zedarvates bundlePropose skills/docs within a token budget using lexical ranking and a conservatively quantized knapsack. Use to shortlist optional context before reading its complete instructions; this does not verify applicability, resolve dependencies or execute skills.
- ▌ Prefix Pruner · zedarvates bundlePrune recognized context sections using stored retention/skip counters. Use to evaluate context reduction on a copy after checking which instructions and information the task requires.
- ▌ Tiered Router · zedarvates bundleEstimate model tiers and token costs, track caller-recorded usage, or inspect in-memory response caching and agent delta helpers. Use when integrating budget-aware routing or analyzing its limits; this module does not execute model calls.
- ▌ Skill Finder · zedarvates bundleRetrieve candidate skills from local SKILL.md files, with optional local-model review of their complete bodies. Use for skill discovery or narrowing a tool shortlist. Results are advisory; choosing an operation also requires checking its instructions, exclusions and task prerequisites.
- ▌ Infra Advisor · zedarvates bundleAudit the local cluster's hardware/software/MCP setup and recommend changes that cut token cost — GPU upgrades, Hailo NPU for vision, moving the inference node to Linux, running Qdrant locally, wiring MCP — with an ASCII cluster diagram. Also provides an auto one-pass audit (directives + infra + code duplication + skills) on the project. Use when the user asks how to reduce token/usage cost via hardware/infra, wants cluster setup tips, an ASCII diagram of their setup, or a quick all-in-one audit.
- ▌ LLM Backends · zedarvates bundleDiscover, audit and use local LLM servers, or configure portable task-specific inference profiles for a user's machines and existing memory. Use for local models, LM Studio, Ollama, hardware-aware setup, local inference benchmarks or native speculative profiles. Not for training draft models or treating two chat APIs as speculative decoding.
- ▌ Meta Harness · zedarvates bundleMeta-Harness orchestre les skills Botte comme des agents interchangeables dans un pipeline gouverné. Planifie → exécute en sandbox → review croisé → applique avec garde-fous et SAFE-EXIT.
- ▌ Capabilities · zedarvates bundleThe system's self-model — a capability registry (scans every SKILL.md into a layered tree SENSE→DECIDE→ACT→REMEMBER→GOVERN→DEPLOY) plus a "curator" that picks the right capabilities for a goal locally. Use to see the whole toolkit as a system/arborescence, to let an agent discover what botte-secrète can do, or as the data the Conductor reads to compose a plan.
- ▌ Control Loop · zedarvates bundleInspect routing telemetry and propose, apply or reset effort-to-tier thresholds. Use when reviewing routing statistics or tuning a shared router; recorded success and savings fields are caller claims, not independently verified outcomes.
- ▌ Botte Proxy · zedarvates bundleInspect and configure the HTTP proxy that transforms chat requests before forwarding to a chosen LLM endpoint. Compression is not reversible through the proxy; remote targets receive forwarded data. Check request compatibility, content integrity and measured provider usage before claiming savings.
- ▌ Prefix Tree · zedarvates bundleStore named prompt prefixes and compare later text with their baseline. Use for inspecting common prefixes or evaluating a baseline-aware prompt transport; this module does not send or reconstruct messages.
- ▌ Auto Router · zedarvates bundleAuto-decide whether a task runs on a LOCAL model or a CLOUD model (DeepSeek, GLM, Nemotron, Grok, Gemma, …) from an automatic effort estimate, and run multi-model fusion (cascade, draft→refine, vote). Use when the user wants automatic local-vs-cloud routing, to add cloud LLM providers, to make local and cloud models collaborate, or mentions effort-based routing, model fusion/ensemble, OpenRouter, DeepSeek, GLM, Nemotron, or Grok.
- ▌ Preflight · zedarvates bundleDeliver shared Parcimonia instructions through a committed project policy and selective local assessment in Claude Code's UserPromptSubmit hook. Use for onboarding contributors, configuring a local project's agent guidance or diagnosing bypassed preflight instructions. This hook does not provide Codex or ChatGPT Work cloud lifecycle support.
- ▌ Trajectory · zedarvates bundleStore solver history and maintain a project-local Quality Compass that learns only from externally verified outcomes. Use when a task needs explainable k-NN recall, qualitative QA labels, a shadow routing baseline, or evidence collection before training a micro-NN.
- ▌ Bootstrap · zedarvates bundleConfigure Botte Secrète in a project with MCP wiring, shared Parcimonia and prefer-local guidance, optional AGENTS.md creation and a setup report. Use to install Botte, onboard a repository or diagnose missing project instructions. Configuration alone does not prove routing, model availability or cost savings.
- ▌ Conductor · zedarvates bundleSelect and order Botte capabilities for a goal, preview their commands, or execute a trusted plan within existing task authority. Use explicit skill plans when local artifact checks, dependencies or checkpointed resumption are needed. Selection is local and lexical; inspect selected skill instructions to confirm applicability before execution.
- ▌ Nn Audit · zedarvates bundleInspect micro-model training-script keywords, metadata keys, test text and source references. These heuristic labels help prioritize review; they do not verify datasets, model accuracy, executed tests or production use. Use to inspect the evidence behind a grounded, synthetic or unknown label.
- ▌ App Test · zedarvates bundleLocal-first GUI/app testing by image matching (SikuliX) — turn a small JSON spec referencing your button images into a runnable SikuliX script and run it locally, using a vision NPU (Hailo-8/10) or local vision model instead of cloud vision. Use when the user wants to test a desktop/game/web app "for real" by clicking buttons, has button images already, or mentions SikuliX, image-matching tests, or local UI tests.
- ▌ Cardinal · zedarvates bundleReview Blue Team audit, fix and optimization results using scoped counter-reviews, or consolidate existing Blue/Red reports. Use when a task needs adversarial review of proposed changes or a report-based confrontation summary.
- ▌ Cluster · zedarvates bundleTreat the homelab/micro-cluster as one schedulable resource — discover every reachable machine and spread cheap work across them (least-recently-used first) so idle boxes get the next task, recovering wasted local capacity instead of paying the cloud. Use to see cluster status, route a task to the idlest machine, or hand a task to a trusted agent on another machine. Also use when the user mentions homelab, cluster, idle machines, load balancing, or distributing work across machines.
- ▌ LLM MCP · zedarvates bundleMCP server that lets Claude Code (or any MCP client) discover and call local LLM servers (LM Studio, Ollama, …) as tools, to offload cheap tasks off the cloud. Use when the user wants to wire local models into their agent, register an MCP server, or have the agent automatically route simple tasks to local hardware.
- ▌ Caveman · zedarvates bundleProduire un prompt de réponse concise ou analyser la taille d'un fichier en lecture seule. Ne compresse pas le fichier et ne mesure pas les économies d'un modèle.
- ▌ Checkup · zedarvates bundleRun the standard project checkup for policy presence, directives, metrics, infrastructure, duplication, security candidates and drift. Use after a component update, when onboarding a project, or when the user requests a checkup or botte doctor.
- ▌ Events · zedarvates bundleProject-local JSONL decision log (.botte/events.jsonl) with rotation — routing, cache hits, escalations, micro-NN outputs. Demo mode, the live dashboard, and session replay read it. Use when you want to see or emit a live feed of routing/cache/escalation decisions, or when building a tool that needs to watch the belt work in real time.
- ▌ Report · zedarvates bundlePersist any audit as a timestamped Markdown and/or HTML file (name + date + time) under .botte/reports/, browsable at any time, and list saved reports. Use when the user wants audits saved to consultable files, a history of audits, or to export a checkup/metrics/infra report as .md or .html.
- ▌ Cache · zedarvates bundleReuse project scan and audit results through a local ProjectCache, or inspect the in-memory FormCache for approximate prompt matches. Use when repeated analysis may reuse a result after checking source freshness and task context.
- ▌
- ▌ Portfolio Sync · zedarvates bundleValidate a private portfolio registry and compare it with a pre-fetched GitHub repository inventory. Strictly read-only, deterministic, standard-library only, no network and no cloud tokens. Use to detect unregistered repositories, stale entries, visibility drift, archive drift, duplicate project IDs, forbidden absolute paths, or accidentally stored credentials before proposing changes to the project cockpit.
- ▌ Completion Proof · zedarvates bundleLocal report-only detector for completion claims without evidence markers (A11). Does not authenticate evidence or block builds.
- ▌ Execution Harness · zedarvates bundleRecord requested versus executed state, evidence, consequences, recovery points, divergent exploration, independent verification, and model-by-harness capability observations without executing consequential actions itself.
- ▌
- ▌ Botte Secrete · zedarvatesUse connected Botte Secrète and Parcimonia for first assessment of research, GitHub/PR review, tool selection, code changes and software or game test planning, and for deterministic statistics, exact file duplicates or local Markdown links. Use without requiring a Botte mention; skip trivial conversation and obvious direct actions. Requires a separately installed compatible runtime.
- ▌ Safe Exit · zedarvates bundleDeterministic bounded-execution guard for agent loops. Use to stop iteration, tool-call, wall-time, repeated-failure, or no-progress loops and to enforce non-escalating authorization plus snapshot gates for destructive actions.
- ▌ Memory Hub · zedarvates bundleCapture, recall and correct sourced project memory through a configured Botte shared-memory service. Use for continuity between agent sessions, work checkpoints and explicit memory corrections.
- ▌ Effect Review · zedarvatesReview a selected operation's likely consequences and its execution evidence, or assess a concrete reuse proposal. Use with Conductor review cues or a task handoff; this supplies a shared review method and does not execute the operation.
- ▌ Clean Cache · zedarvates bundleClean development caches (.botte-cache, .pytest_cache, __pycache__, .mypy_cache)
- ▌ Fix · zedarvates bundleList a project's correctable issues — confirmed dead code, duplication, stale directive references — each with a tokens·model·money·time cost estimate and a total. Plan-only by design (never edits code automatically). Use when the user asks what's worth fixing and what each fix costs.
- ▌ Demo · zedarvates bundleLive ANSI dashboard of the belt's decisions — routing, token savings, micro-NN outputs, escalations, cache hits — either a built-in scripted scenario (no LLM, no network, works on a bare machine) or tailing a real project's event log. Use when the user wants to see/demo what the routing belt is doing, record a README GIF, or watch live decisions while an agent works.
- ▌
- ▌ Bench · zedarvates bundleReproducible token/cost benchmark — runs a fixed task corpus through the real auto_router decision logic and compares it against a "no routing, everything to cloud STANDARD" baseline. Turns the README's savings claim into a checkable number instead of an assertion. Use when the user wants proof of token savings, a benchmark for a PR/README, or numbers to back an integration pitch (e.g. Hermes).
- ▌ Cwe Kb · zedarvates bundleLocal CWE knowledge base (RAG) to enrich and de-noise security findings — match a finding (or any text) to the relevant weakness by exact id or local-embedding similarity, and attach the weakness name, description, and concrete mitigation. Deterministic, offline, 0 cloud tokens. Use to explain a CWE, find the likely weakness for a code snippet, or enrich taint/security findings with "why + how to fix".
- ▌ Docgen · zedarvates bundleGenerate documentation with a local model drafting and the cloud only refining (0 cloud tokens for the draft), plus a local session review that summarises what a work session did. Use to write README/module docs/CHANGELOG/guides/ADRs cheaply, or to produce end-of-session notes/learnings from a transcript.
- ▌ Ingest · zedarvates bundleLocal-first web scraping and source ingestion — fetch a URL, extract clean text locally (0 cloud tokens), optionally structure it with a local model, and store it in a Qdrant collection (the second-brain "foundation") for later recall. Use for /web-scraping, /ingest-source, building a knowledge foundation, or recalling past ingested sources. Web fetch + Qdrant over stdlib HTTP.
- ▌
- ▌ Trends · zedarvates bundleTrack a project's audit metrics over time (directive score, duplication, LOC, always-on cost, fix count) and show the change since the previous run. Use to see whether the project is getting healthier/cheaper across audits.
- ▌ Metrics · zedarvates bundleCost-focused project metrics, broken down per component — LOC by language and component, duplicate-function groups, directive health, always-on context cost (CLAUDE.md tokens × turns), local-routing posture, skill-search tokens avoided, and the audit's own (near-zero) cost. Use when the user wants to quantify a project's token/cost profile, see LOC/health per component, or measure what the toolkit saves.
- ▌ Solvers · zedarvates bundleDeterministic combinatorial solvers in stdlib — balance work across workers/backends (assignment, LPT), pack items under a capacity (bin-packing, FFD), and order plan steps under dependencies into a sequence + parallel waves (DAG topological scheduling, cycle-detecting). 0 cloud tokens, repeatable. Use to spread cluster work, pack tasks under a budget/capacity, or order a plan's steps — instead of asking an LLM to figure out the assignment/order.
- ▌ Dashboard · zedarvates bundleGenerate one self-contained, timestamped HTML dashboard of the system's cost picture — routing savings (control loop), metric trends, current metrics, and the cost of outstanding fixes. Also renders as a live ANSI terminal view (--tui, --watch) and serves a live HTTP API (api.py). Use when the user wants a single visual view of cost/savings/health over time, or a live terminal view they don't have to open a browser for.
- ▌ Statusline · zedarvates bundleOne-line summary of the belt's session activity (tokens saved, cache hits, local/cloud split, escalations) for a terminal statusline — Claude Code's statusLine hook, tmux, or any shell prompt. Reads .botte/events.jsonl. Use when the user wants a persistent, passive view of savings while they work, or asks to set up a statusline.
- ▌ Agent Cache · zedarvates bundleCache les réponses des agents pour skipper l'exécution quand l'output est prédictible. Use when you want to cut 10-15% by avoiding redundant agent runs.
- ▌ Auto Memory · zedarvates bundleMemory as a learnable skill — store, recall, compress, and consolidate agent memories. Inspired by Stanford AutoMem.
- ▌
- ▌ MCP Gateway · zedarvates bundleMCP Gateway — expose toutes les skills Botte comme outils MCP. Découverte automatique, schémas d'entrée, transport stdio. Compatible Claude Code, Codex, Cursor, et tout client MCP.
- ▌
- ▌
- ▌ Docs Steward · zedarvates bundleScoped documentation map for multi-component projects (server + client + tools + …). Detects components, classifies every doc as global vs component-scoped, and produces a per-component index (DOCS.md) listing local docs + links to the relevant global docs — so an LLM coder bounded to one component loads only its scope, not every other component's documentation. Frames token cost (full project docs vs scoped load) and treats .md as LLM-facing, .html as human reference. Use when a project has several components and you want to cut the docs an agent must read, or asks how to organise docs for a monorepo.
- ▌ Fast Context · zedarvates bundleFastContext Agent — exploration repo déterministe. Parse une requête d'exploration → READ/GLOB/GREP ciblés → rapport compact (fichier:ligne:score). Remplace 56% des appels LLM de type "read/search" par des opérations stdio à ~5ms, 0 token. Use when the agent needs to understand a codebase, find patterns, locate imports, or gather context without an LLM call.
- ▌ Monte Cristo · zedarvates bundleIndependent strategic outsider above the blue and red teams. Use for architecture resets, high-cost research decisions, inherited assumptions, and stalled adversarial reviews; do not use for routine review or autonomous mutation.
- ▌
- ▌ Asset Quality · zedarvates bundleEvaluate generated images, textures, meshes, animations, and Godot packages with deterministic gates followed by an explainable family-local k-NN memory.
- ▌
- ▌
- ▌
- ▌ Harness Delta · zedarvates bundleVérification différentielle — ne vérifie que les sections modifiées.
- ▌ Hermes Bridge · zedarvates bundleExpose routing, local chat, fusion, skill search, infrastructure advice, and privacy-safe QA run manifests to Hermes-Agent or another OpenAI-function-calling framework, plus a one-call MCP config generator. Use when connecting botte-secrète's routing and evidence contracts to another agent framework.
- ▌ Hf Provenance · zedarvates bundleCompare a downloaded Hugging Face JSON-weight snapshot with the authoritative Botte micro-NN source. Produces SHA-256 inventories and blocks publication on missing, extra, changed, invalid, or unproven model files. Offline and dependency-free.
- ▌
- ▌
- ▌
- ▌ Cost Estimator · zedarvates bundleEstimate what a task or a fix will cost — tokens, model/tier, money ($), and wall-time — using the tiered cost model. Use whenever the user wants to know the cost of a correction, audit, or task before running it, or to compare local vs cloud cost.
- ▌ Context Windows · zedarvates bundleFenêtres de contexte pour boucles rétroactives — charge seulement les deltas.
- ▌ Decision Ladder · zedarvates bundlePonytail-inspired YAGNI enforcement — climb the decision ladder before writing any code. stdlib → regex → existing module → new code.
- ▌ Migration Audit · zedarvates bundleDeterministically verify that a framework, protocol, dependency, or build migration actually removed legacy paths and markers before VALIDATOR runs. Returns PASS, FAIL, or UNCERTAIN with bounded path-only evidence; never executes project code.
- ▌ Prompt Improver · zedarvates bundleRewrite a rough prompt into a professional, structured prompt (role, context, task, instructions, constraints, output format, success criteria) using a LOCAL model — 0 cloud tokens — with markdown or strict JSON prompt output. Use when the user wants to improve/boost/structure a prompt, mentions "/p-amelioration", "prompt pro", "JSON prompt", prompt engineering, or wants a reusable prompt template built locally.
- ▌
- ▌ Context Profiler · zedarvates bundleMeasure a project's always-on prefix (agent directives + core rules + MCP tool schemas + skill catalogue) in tokens and as a % of small local-model windows (64k/128k/256k), with a concrete reduction plan (lazy tool loading, on-demand skill search). Use to see how much of a modest machine's usable context is spent before any real work, and how to shrink it so weaker machines can run local LLMs usably.
- ▌ Directives Audit · zedarvates bundleAudit a project's AI-agent guidance files — CLAUDE.md, AGENTS.md, .cursorrules, copilot-instructions.md, GEMINI.md, intent docs and specs, in markdown, text or HTML. Use when checking whether a repo has agent instructions, whether they are healthy/consistent, or before an agent starts work on an unfamiliar codebase. Also use when the user mentions CLAUDE.md, AGENTS.md, agent rules, "does this project have instructions", or stale/oversized directive files.
- ▌ Security Scanner · zedarvates bundleScan Python skills and MCP servers for malicious code — dangerous imports, network exfiltration, filesystem abuse, subprocess injection, obfuscation, crypto weakness, environment leaks, and supply-chain attacks. Use when auditing code for security, running pre-commit hooks, or preparing CI/CD security gates.
- ▌ Nlp Deterministic · zedarvates bundleClassify and extract from text WITHOUT an LLM — intent classification (lexical overlap + local embedding), entity extraction (regex/gazetteers for urls/emails/ips/paths/env vars/flags/numbers), and stopword-filtered keyword frequency. Deterministic, instant, 0 cloud tokens. Use instead of asking a model to classify text or extract entities, and as the routing/intent layer that keeps cheap language decisions off the LLM.
- ▌