stashai
CLI for Stash — shared memory for AI coding agents
Decision gist · record as of 2026-08-14
Yes, if you run coding agents regularly and want to avoid duplicating work across sessions. Stash is actively maintained, has low install friction, and solves a real problem—agents can only access transcripts from their current machine by default. MIT license is permissive. Beta status means the API may shift, but the core functionality (session recording, multi-source search, MCP integration) is documented and in use. Start with `stash signin` to see if the setup and agent detection work for your toolchain.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.11 or later.
- First run (`stash signin`) opens a browser for authentication and walks through setup including session recording opt-in and agent selection.
- Low install friction with a pure Python wheel and six runtime dependencies (typer, httpx, rich, questionary, mcp, filelock).
License · maintenance · safety
MIT (permissive) — MIT license (permissive) means you can use, modify, and distribute Stash freely in commercial and private projects with minimal restrictions — only attribution and license inclusion required.
last release 2026-08-10 (4 days) · last repo commit 2026-08-14 · 313 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 98,069 downloads/mo, #13,110 on PyPI
Alternatives
Verify before relying
# Install via uv
uv tool install stashai
# Authenticate and set up
stash signin
# Query your knowledge base
stash vfs ls /
stash search "your query here"- Whether the 49% speedup claim for Claude Code is reproducible across different codebases and agent configurations.
- Exact scope and latency of federated search across all connected sources in a typical multi-source setup.
- Self-hosted deployment requirements (Docker, resource footprint, maintenance burden) beyond the docker-compose example.
- Data retention, backup, and disaster-recovery guarantees for agent sessions and imported knowledge.
- Whether MCP server (~70 tools) is fully documented and stable across different agent implementations.
What it is and what it does
Stash is a knowledge-base platform designed to give coding agents persistent, cross-session memory. It ingests agent session transcripts, files, and data from a wide range of sources—GitHub repos, Google Drive, Gmail, Slack, Notion, Linear, Jira, Asana, Granola, and others—and stores them in a searchable, agent-native format. Agents can query this knowledge base via a CLI, MCP server, REST API, or virtual filesystem shell, so every agent run has context about every previous session you've created.
The package provides a CLI (`stashai`) that handles authentication, session recording, source integration, and agent setup. It also runs an MCP server with approximately 70 read/write tools, allowing agents to search, read, and write to the knowledge base in real time. A built-in curator can automatically compile new sessions and files into linked wiki pages, and a Skills system lets you package and share reusable knowledge folders. The tool is designed for teams running long-lived or autonomous coding agents, where duplicating work across sessions is costly and context loss is a real problem.
Use it for
- Avoid re-solving the same bug: ask your agent if it has tackled a memory leak or similar issue before and retrieve the transcript of what was attempted.
- Track work across agents and machines: maintain a persistent log of what every coding agent has done, not just sessions on the current machine.
- Build living documentation: let agents write and update ADRs, design notes, and runbooks as they work, keeping them current without manual sync.
- Recover lost context: search for why a decision was made (e.g., why a timeout was set to 30s) without relying on git history or scattered notes.
- Share knowledge between teammates: publish a Skill (a folder with pages and tables) so other agents and team members can fork or follow it.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you run coding agents regularly and want to avoid duplicating work across sessions.
Stash is actively maintained, has low install friction, and solves a real problem—agents can only access transcripts from their current machine by default. MIT license is permissive. Beta status means the API may shift, but the core functionality (session recording, multi-source search, MCP integration) is documented and in use. Start with `stash signin` to see if the setup and agent detection work for your toolchain.
Install
stashai on PyPI
Before you install
Low install friction with a pure Python wheel and six runtime dependencies (typer, httpx, rich, questionary, mcp, filelock). Active maintenance: last commit 2026-08-14, version 0.1.364 released 2026-08-10, and repo shows 313 stars. Beta status (Development Status :: 4) is appropriate for a young project in active development.
Requires Python 3.11 or later. First run (`stash signin`) opens a browser for authentication and walks through setup including session recording opt-in and agent selection.
License in practice
MIT license (permissive) means you can use, modify, and distribute Stash freely in commercial and private projects with minimal restrictions — only attribution and license inclusion required.
Quickstart
# Install via uv
uv tool install stashai
# Authenticate and set up
stash signin
# Query your knowledge base
stash vfs ls /
stash search "your query here"Verify before relying
- Whether the 49% speedup claim for Claude Code is reproducible across different codebases and agent configurations.
- Exact scope and latency of federated search across all connected sources in a typical multi-source setup.
- Self-hosted deployment requirements (Docker, resource footprint, maintenance burden) beyond the docker-compose example.
- Data retention, backup, and disaster-recovery guarantees for agent sessions and imported knowledge.
- Whether MCP server (~70 tools) is fully documented and stable across different agent implementations.
Package facts
| License | MIT permissive |
| Python support | Supports the current Python release >=3.11 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 6 packagestyperhttpxrichquestionarymcpfilelock |
| Maintenance | Actively maintained 4 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 98,069 / month, #13,110 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: DevelopersLicense :: OSI Approved :: MIT LicenseProgramming Language :: Python :: 3Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Topic :: Software Development :: Libraries :: Python Modules |
Evidence: stashai-0.1.364-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “persistent context for coding agents”
- stashaiStash is a CLI and MCP server that creates a persistent,…
- memoriMemori is a Python SDK that automatically captures and recalls…
- cogneeCognee builds a self-hosted knowledge graph from ingested data,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Python Modules packages
Converts domain names between Unicode and ASCII-compatible encoding (Punycode) according to IDNA 2008 and Unicode Technical Standard 46, with security validation and broader script coverage than the standard library.
Install it if you work with internationalized domain names, need to validate domains, or use HTTP clients that depend on it transitively.
Setuptools is a Python build backend and package management tool that handles building, distributing, and installing Python packages, including support for C/C++ extension modules.
PyYAML parses and emits YAML 1.1 data format, enabling serialization and deserialization of configuration files and Python objects to and from human-readable YAML text.
Pydantic validates Python data structures against type hints, coercing and checking input at runtime to ensure it matches a declared schema.
Provides reusable metadata objects for use with PEP-593 `typing.Annotated` to express common constraints like bounds, collection sizes, and predicates on types.
Install it if you use or build libraries that need to express type constraints in a standardized, inspectable way—or if you want to annotate your own types with…
Provides runtime tools to inspect and introspect Python type annotations, enabling programmatic examination of type hints at execution time.
See also cognee · mindroom · reme-ai · google-agents-cli · openviking · memsearch · mem0ai · apm-cli · agent-utilities · semble