claude

Knowledge Forge

Turns "explain X" moments into permanent, linked, verified markdown notes in an Obsidian vault. A zero-model-call Go static-analysis core (recall, drift, quality gates) with an optional four-tier LLM layer on top.

3 on GitHubby Samirv0.1.2MITupdated 2 months ago
/

Plugin forge from marketplace knowledge-forge in mimir45/Knowledge-Forge. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (6)

  • forge-check

    Use when the user asks to run the weekly vault health check — "run forge check", "what needs attention in the vault", "give me the weekly rollup", "any drift/duplicates/orphans to look at", or on a scheduled/cron-style invocation of the weekly job. Not for a single explanatory question (that's `forg

  • forge-dataset-stats

    Use when the user asks what training data the vault has accumulated — "how much training data do I have", "can I fine-tune on this yet", "show dataset stats", "what's in d1/d3", "is there enough for a LoRA". Not for whether the vault is paying off day to day (that's `forge-stats`, which reads the te

  • forge-export-dataset

    Use when the user wants to get their captured training data out — "export the dataset", "give me the d3 pairs", "export for fine-tuning", "dump the routing data as CSV", "anonymize and export". Not for asking how much data there is (that's `forge-dataset-stats`, which writes nothing) and not for scr

  • forge-init

    Use when the user wants to set up Knowledge Forge for a vault they have not configured yet — "set up forge", "install knowledge forge here", "onboard this vault", "run forge init", or the first time `forge recall`/`forge check` fails because no config exists. Not for changing an existing config (edi

  • forge-stats

    Use when the user asks whether Knowledge Forge is paying off — "is the vault helping", "what am I asking about most", "show forge stats", "what gaps are there", "how's the hit rate", "am I saving time with this". Not for a single explanatory question (that's `forge`), not for the weekly health rollu

  • forge

    Use when the user asks an explanatory question about a technology, concept, or integration — "how does X work", "what is X", "explain X", "how do I integrate X with Y", "what's the difference between X and Y", "best practices for X", "why does X do Y". Checks the vault before researching, then answe

Subagents (4)

  • forge-codebase-scout

    Finds how a topic is actually used in the current repo — file:line examples, local conventions, project-specific config — as the differentiator between a generic note and one grounded in this codebase. Read-only, no Bash.

  • forge-librarian

    Merges/links a gated note into the vault, populates code_refs, rebuilds the index, and commits with the Forge-Write trailer. Runs only after forge gate has already passed a draft — never a gate bypass.

  • forge-researcher

    Answers N research questions about an external topic (a library, protocol, pattern) from authoritative web sources, returning findings with dated citations. Read-only — never writes to the vault or the repo.

  • forge-verifier

    Spot-checks a draft's claims and code snippets — compiles snippets via forge verify-code, cross-checks version-specific claims against sources — and returns pass/fail per claim plus a confidence assignment. Never runs a compiler by hand.

Hooks (4)

  • SessionStart

    "${CLAUDE_PLUGIN_ROOT}"/hooks/session-context

  • UserPromptSubmit

    "${CLAUDE_PLUGIN_ROOT}"/hooks/user-prompt-intent

  • SessionEnd

    "${CLAUDE_PLUGIN_ROOT}"/hooks/session-end-capture

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}"/hooks/post-tool-cache-source

View source on GitHub

Similar plugins

Statically analyze a local JS/Electron app you're authorized to study, with file+line evidence, and build an original reimplementation of what you find. Experimental. Phase 1: JS/Electron static analysis. Phase 2: native ELF/PE/Mach-O container inspection. Phase 3: .NET assembly metadata. Phase 4: EVM bytecode inventory. Phase 5: CUDA fatbin, cubin and PTX inventory. Parses, never executes. Dynamic analysis is a future phase, not yet implemented.

1 skill · 1 command · 1 agent

ruvnet/ruflo 81,589

Generate natively editable PPTX from PDF / DOCX / URL / Markdown — real DrawingML shapes, text boxes, charts, and animations. Note: run `pip install -r requirements.txt` inside the installed plugin directory for the Python post-processing scripts to work.

1 skill

hugohe3/ppt-master 76,375

obsidian

claude

Claude Skills for Obsidian

6 skills

kepano/obsidian-skills 51,587

gitnexus

claudecodex

Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.

12 skills · 1 MCP server · hooks

abhigyanpatwari/GitNexus 48,482

carta-crm

claude

Manage the Carta CRM conversationally — search, add, update, and enrich investors, companies, contacts, deals, notes, and fundraisings via the Carta CRM MCP Server.

28 skills · hooks

anthropics/claude-plugins-official 47,076

Earnings call and filings to model update to note draft

6 skills · 1 agent

anthropics/financial-services 41,259

Maintain Knowledge Forge?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted