jamesob
Featured
local-llm
A comprehensive guide for building and configuring a high-end local machine to run state-of-the-art LLMs, with detailed hardware choices, BIOS tuning, and Docker-based model serving.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It exposes the internal system-skill architecture of a production coding agent, including self-improvement, self-knowledge, and taste, making it a useful reference for understanding how modern AI coding tools are structured.
Who it is for
- Developers building or extending AI coding agents
- Researchers studying agent skill and prompt architecture
- Sarvam Code users who want to understand its built-in capabilities
- Engineers interested in OCR, recursive LM, and self-improvement patterns
Use cases
- Studying how built-in agent skills are organized and progressively disclosed
- Adopting similar SKILL.md/references/scripts patterns in another agent project
- Learning practical workflows for document digitization, PDF rasterization, and map-reduce summarization
- Exploring Sarvam Code's configuration surface and self-improvement loop through its reference files
Strengths
- Contains exact skill content extracted from a specific version, preserving real structure
- Covers a broad range of capabilities: OCR, PDF rendering, recursive summarization, self-improvement, configuration, and design taste
- Uses a familiar progressive-disclosure layout that makes the skill tree easy to navigate
- Includes notable reference files such as the Harbor backend explanation and the config.toml knob catalog
Considerations
- The snapshot is tied to v0.38.0 and may become stale as Sarvam Code evolves
- All skill content is copyrighted by Sarvam AI and marked for study, so reuse is limited
- Only the bundled system skills are included; other CLI components or user-created skills are not covered
README quick start
Sarvam Code — bundled system skills
The six system skills that ship inside the Sarvam Code CLI (sarvam-code,
v0.38.0, a codex-rs fork), extracted verbatim from ~/.sarvam/skills/.system/.
These are the agent's built-in capability + self-knowledge tree — the same
progressive-disclosure SKILL.md + references/ + scripts/ layout Claude Code
and codex use.
| Skill | What it does |
|---|---|
| document-digitization | OCR an uploaded document to markdown/HTML across 22 Indian languages + English (upload → digitize_document w/ language code → poll → fetch). |
| document-rasterization | Render a local PDF's pages to PNGs so a vision model can read them ("summarize pages 3–10", "what's on page 5"). |
| rlm-mode | "Recursive LM" mode for input-bound work — label/classify/aggregate/dedup/summarize over large sets by holding data in a program (run_program) and dispatching recursive llm() calls over slices, then combining. |
| self-improvement | The governor: black-box optimization of the agent harness (config + skills + prompt) around a fixed model, evaluated on the Harbor backend or in-process replicas, scored on a (reward, cost) frontier. |
| self-knowledge | Authoritative reference tree for operating/configuring/extending Sarvam Code itself — config.toml, models, features, memory, compaction, config layers, hooks, MCP, sub-agent roles, slash commands. |
| taste | The code-design/review bar — loaded before choosing a design for changes that carry design decisions (types, refactors, API changes, multi-file edits). |
Notable references
self-improvement/references/harbor.md— how Sarvam runs the self-improvement loop on Harbor (harborframework.com): per-dataset config, oracle gate, ATIF trajectories, the two config layers, the document-eval failure taxonomy.self-knowledge/references/config/*.md— the fullconfig.tomlknob catalog.rlm-mode/scripts/summarize_map_reduce.ts— the map-reduce driver.
Extracted for study. All skill content is © Sarvam AI.
Description
The 6 system skills bundled inside the Sarvam Code CLI (v0.38.0), extracted for study
Related repositories
Similar projects matched by category, topics, and programming language.
0xwilliamortiz
Featured
openclaude-improved
OpenClaude is an open-source CLI coding agent that runs on any platform and supports a wide range of LLM providers, offering the same tools and workflows as Claude Code.
slvDev
Featured
esp32-ai
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.