Search and retrieve security commands, exploit scripts, and methodology from your pentest writeups and shell history. CLI + MCP server for AI assistants.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"command-vault-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# Command Vault (also MCP) > *"What was that certipy command I used for ESC8?"* > *"How did I exploit that shadow credentials thing again?"* Command Vault indexes **commands**, **scripts**, and **prose** from your penetration testing writeups and shell history into a searchable database with full context — what tool, what technique, which box. MCP-ready for AI assistants. ## Features - **Command search** — Ranked FTS across command text and its context, with tool/category/tag filters - **Prose search** — Search section-aware evidence, including methodology and fenced log/XML text; read source context by reference - **Script search & retrieval** — Find detected scripts by language/library and retrieve their indexed code by ID - **Ranked fallback** — Multi-word queries try AND first, then BM25-ranked OR if no AND matches exist; fallback is not a relevance guarantee - **Shell history** — Index `~/.zsh_history` or `~/.bash_history` with deduplication and selected redaction patterns; dates stay unknown when absent - **Tag filtering** — Search by extracted `#hashtags`; all requested tags must match - **Smart categorization** — 200+ tool names mapped to categories (recon, AD, web, privesc, etc.); unknown names fall back to `misc` - **Template generation** — Heuristic placeholders for recognized IPs, lab domains, and credential arguments; not comprehensive secret removal - **Multiple writeup types** — Boxes, challenges, and Sherlocks with unified or legacy directory modes ## Installation Requires Python 3.11+. [uv](https://docs.astral.sh/uv/guides/projects/) is recommended for the source installation; it is not required by the installed runtime. To install from source: ~~~bash git clone https://github.com/x746b/command-vault.git cd command-vault uv sync --frozen source .venv/bin/activate ~~~ If an existing checkout uses `.venv-v2`, use `UV_PROJECT_ENVIRONMENT=.venv-v2 uv sync --frozen` and activate `.venv-v2/bin/activate` instead. The examples below assume the sele…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.