BladesvsCersei
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Blades
go-kratos · Agent framework
OSS
Panel
6.31 spec wins
- Reliability
- 6.5
- Usefulness
- 5.0
- Cost
- 7.5
- Longevity
- 6.2
“Named after the weapons Kratos carries in God of War, which is a lot of commitment for a package you import.”
Cersei
Adib Mohsin · Agent framework
OSSMCP
Panel
6.28 spec wins
- Reliability
- 5.7
- Usefulness
- 6.0
- Cost
- 8.3
- Longevity
- 4.7
“It reads Claude Code's skills files and writes Claude Code's session files, so the migration path is helpfully signposted in one direction.”
Spec by spec
| Spec | Blades | Cersei |
|---|---|---|
| Architecture | ||
| Category | Agent framework | Agent framework |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | No | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | No | Yes |
| Multi-file edits | No | Yes |
| Git operations | No | YesAbstract ships /commit, /review and /diff slash commands over the working tree. |
| Browser control | No | YesWeb tools are among Abstract's 34 built-in tools; no browser automation loop is documented. |
| Sandboxed execution | No | No |
| Multi-agent orchestration | No | Yes |
| Headless / CI mode | No | YesAbstract offers --no-permissions with NDJSON output as a CI mode. |
| Models | ||
| Backbone | OpenAI, Gemini, DeepSeek | Anthropic, OpenAI, Ollama, Azure OpenAI, vLLM |
| Bring your own model | YesThe ModelProvider abstraction is documented as supporting custom models. | Yes |
| Local models | No | Yes |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | MIT |
| GitHub stars | 817 | 460 |
Which one would each critic pick
| Critic | Blades | Cersei | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.3 | 6.8 | Cersei — Pick it when the agent belongs inside your application; pick a terminal agent when you only want something to edit the repository while you watch. |
| El Crítico | 5.5 | 5.8 | Cersei — The design is derived from another vendor's agent, and the crate is installed from a git reference rather than a released version, so there is nothing to pin against upstream drift. |
| El Profesor | 6.8 | 6.0 | Blades — Middleware as a core concept is the interesting design decision; chain-of-thought is listed beside it as a capability, which is a property of prompts, not of a library. |
| La Inversora | 6.3 | 5.8 | Blades — This belongs to an established open framework project rather than a startup, and 811 stars is a subproject's audience, not a market. |
| La Jefa | 5.8 | 5.3 | Blades — Nothing to license across sixty seats because there is nothing to license: this is an import, and the operating cost is whatever my team builds on top of it. |
| El Hacker | 7.3 | 7.5 | Cersei — MIT, my key or no key at all, and a local vLLM or Ollama server is a first-class provider rather than a compatibility footnote. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.