Memharness vs Skill Federation

Memharness and Skill Federation are both agent frameworks tracked by AIDiveForge. Below is a side-by-side comparison of pricing, capabilities, platforms, and ownership — sourced from each tool's live website and verified before publishing.

Memharness

The core premise is storing facts, not strings, with two independent time axes: when something became true in the world and when the agent learned it — so querying past agent states is a real query, not archaeology through logs. Everything lives in a single SQLite file, which means the storage layer makes zero LLM or network calls and stays auditable. Recall combines hybrid vector search and full-text search with a source-staleness signal, so older or superseded sources rank down automatically. Where it breaks: the SQLite backend is a hard ceiling for teams expecting distributed writes or high-concurrency production deployments. Teams hitting that ceiling will need to treat memharness as a pattern to port, not a service to scale horizontally.

Skill Federation

Skill Federation runs locally on your machine and connects to a catalog of over 100,000 vetted skills. When an agent hits a gap, it surfaces matches in milliseconds — each one license-checked, security-scanned, and provenance-tracked — then waits for your approval before installing into .claude/skills/. The benchmark evidence from the vendor is specific: a bare Claude Code agent solves 17.5% of SkillsBench tasks; with Skill Federation retrieving the top match, that climbs to 22.8%, roughly closing 27% of the gap to a hand-crafted ideal skill. The privacy boundary is narrow by design — only an abstract wish crosses the wire, never your code, plan, or outputs. The hard ceiling is integration breadth: Claude Code is supported, with Codex, Cursor, and Gemini listed as coming.

Attribute	Memharness	Skill Federation
Pricing	Free	Free
Free trial	No	No
Open source	Yes	Yes
Has API	Yes	No
Self-hosted option	Yes	Yes
Platforms	SQLite, MCP	Node.js, Python, cross-platform via curl
Pros	Bi-temporal storage tracks both world-time and agent-learn-time independently, so you can reconstruct exactly what the agent believed at any past moment — which means post-incident reviews and compliance audits have an actual record to query instead of inferring from logs. Provenance-scoped deletion lets you remove all facts derived from a specific source in one operation, so GDPR takedown requests or source revocations do not require a full memory wipe that destroys unrelated facts. The storage layer makes zero LLM or network calls, so memory reads and writes have no latency dependency on external APIs and no token cost — which means memory operations do not blow your inference budget. Hybrid vector-plus-full-text recall with a built-in staleness signal means older or superseded sources rank lower automatically, so the agent surfaces the most current relevant facts without you writing custom re-ranking logic. MCP exposure and a self-hosted SQLite backend mean the tool drops into any agent stack that speaks MCP without requiring a separate managed service, so you retain full data ownership and avoid a vendor dependency in the memory layer.	Two-scanner security vetting at catalog ingestion rather than at install time, so you are never pulling live from an unreviewed repo and your team avoids the malware-by-star-count gamble. License class and provenance shown before every install, which means teams with compliance requirements can audit what skills entered the codebase without reconstructing that history after the fact. Retrieval triggered automatically when the agent hits a gap, so the agent does not require manual skill reminders at the start of every session — a friction point the vendor explicitly benchmarks against. All execution happens on your machine with only an abstract wish transmitted, so codebases, plans, and outputs stay local even when skill search is delegated to an external catalog. Open-source and self-hostable, which means teams that need an air-gapped or fully controlled registry can run their own instance rather than depending on a hosted endpoint.
Cons	SQLite is a single-writer database: the moment two agent processes attempt concurrent writes — a parallelized pipeline, a multi-worker deployment, any architecture where more than one process holds the file — writes will collide or block. Teams with concurrent-write requirements either serialize all memory operations through a single process (adding a bottleneck) or abandon memharness for a Postgres- or Redis-backed alternative. The project has 2 stars and 1 fork on GitHub at time of curation, with 19 commits and no open issues, which means community-sourced debugging, third-party integrations, and production war stories are essentially nonexistent. Teams that hit an edge case are reading the source, not a Stack Overflow thread. There is no built-in access control or multi-tenant isolation: if multiple agents or users share the same SQLite file, provenance-scoped deletion could become a liability rather than a feature — one delete call wipes facts for every tenant who learned from that source. Teams building multi-user applications will need to implement per-user database files or a sharding layer before going to production.	Claude Code is the only documented supported integration; teams running Cursor, Codex, or Gemini as their primary agent tool cannot use Skill Federation in its current state — those integrations are listed as forthcoming with no committed timeline on the vendor page. The benchmark ceiling exposes the retrieval model's limit: even the top retrieved skill closes only 27% of the gap to a hand-crafted ideal skill, meaning tasks that require precise, purpose-built skills will still be partially solved at best — teams with narrow, specialized workflows will hit this ceiling faster than teams with general-purpose tasks. No API is available, so teams that want to integrate skill retrieval into a custom agent pipeline or CI workflow cannot call Skill Federation programmatically — teams needing that surface will need to build their own retrieval layer, at which point Skill Federation's catalog is no longer in the loop.

Bottom line

Only Memharness exposes a public API. Choose based on which difference matters most for your workflow.

Frequently asked questions

What is the difference between Memharness and Skill Federation?

Memharness is Free and open source, while Skill Federation is Free and open source. Compare pricing, free trial, API, platforms, and pros/cons in the table above on AIDiveForge.

Is Memharness better than Skill Federation?

It depends on your workflow. Use the side-by-side attributes (pricing, open source, API, self-hosted, platforms) to decide. AIDiveForge does not rank a universal winner — we publish verified facts so you can choose.