Semble
Dev tools
code search for agents
SHOW HN2026
445points
151comments
#64of 4,366 in 2026
Top 2%of 2026 by points
How HN took it
CuriousMostly questions and neutral comments
16 positive, 43 neutral or asking, 3 mixed, 14 critical, out of 76 comments about the project. Thread score +0.02.
Performance: 4 for, 6 againstIdea: 5 for, 5 againstExecution: 0 for, 2 againstDocs: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Fast indexing and queries, good retrieval quality versus existing tools
From 2 comments
So about a year ago I wrote my own attempt at something like this using vector indexing and BM25 (the latest version uses CocoIndex, I had a custom coded solution using ChromaDB…
bluejay2387 on HNSeems like a cool idea so I decided to play with it a bit. The test I ran was in the browsercode (https://github.com/browser-use/browsercode) repo with the following prompt:…
aadishv on HN
Useful for humans as well as agents
From 1 comment
Semantic code search seems like a useful tool for a human too. Not just for agents.
singpolyma3 on HN
Objections
Token savings may not translate to lower cost or faster end-to-end agent work
From 3 comments
I did some evals with pi and GPT 5.5. I tested RTK on / headroom on / both on / both off (all with the standard pi system instructions and no AGENTS.md). I forget the exact tests…
esperent on HNWhat I have personally observed with such tools is that they make the AI's dumb, similar to how it makes coders dumb when relying more on AI tools. These agentic AI's are already…
freakynit on HNSeems like a cool idea so I decided to play with it a bit. The test I ran was in the browsercode (https://github.com/browser-use/browsercode) repo with the following prompt:…
aadishv on HN
Agents still fall back to ripgrep and do not trust results
From 2 comments
I'd be interested in seeing actual agent benchmarks (eg CC or Copilot CLI with grep removed and this tool instead). For example, I have explored RTK and various LSP…
jerezzprime on HNFeedback: codex-cli hangs when calling this through the MCP. The semble process even sticks around as a zombie, forever stalled out. No idea why, logs have nothing. When called…
cityofdelusion on HN
Probabilistic search may miss important results
From 1 comment
fantastic token savings and performance... but unlike grep it's probabilistic search on search terms. is that an issue? the tiny model might not surface something important
onoesworkacct on HN
98% savings baseline is unclear
From 3 comments
> uses 98% fewer tokens than grep So are we supposed to believe that grep is so wasteful that models are reading 98% useless garbage every time they call it? Either this claim is…
pu_pe on HNgrep doesn't need tokens, so what is 98% fewer than zero?
ludicrousdispla on HNThe savings are calculated against the assumption that agents read the matched files in their entirety. In my experience, they are smart enough to use grep to bring up a few…
PufPufPuf on HN
Requests
Agent-level benchmarks and end-to-end savings
From 3 comments
I'd be interested in seeing actual agent benchmarks (eg CC or Copilot CLI with grep removed and this tool instead). For example, I have explored RTK and various LSP…
jerezzprime on HNWould be nice to see the actual % token/time saving across end-to-end coding sessions over these time periods https://github.com/MinishLab/semble#savings My observation is that…
cagz on HNI did some evals with pi and GPT 5.5. I tested RTK on / headroom on / both on / both off (all with the standard pi system instructions and no AGENTS.md). I forget the exact tests…
esperent on HN
Better install and usage docs
From 2 comments
The instructions on how to install and use it could use some work. I did eventually install it. Will try it later and report back.
zhxiaoliang on HNFeedback: codex-cli hangs when calling this through the MCP. The semble process even sticks around as a zombie, forever stalled out. No idea why, logs have nothing. When called…
cityofdelusion on HN
Comparison with other tools such as LSPs, colgrep, codebase-memory-mcp and cs
From 4 comments
Better than grep obviously, but how does this compare to existing LSPs?
AussieWog93 on HNHow does this compare with colgrep? https://github.com/lightonai/next-plaid/tree/main/colgrep
nextaccountic on HNWould this replace something like codebase-memory-mcp[1] or improve when both is being used? [1] - https://github.com/DeusData/codebase-memory-mcp
_ink_ on HNInteresting. I too have been working in this space, though I took a different approach. Rather than building an index, I worked on making a "smarter grep" by offering search over…
boyter on HN
Fix for codex-cli MCP hang
From 1 comment
Feedback: codex-cli hangs when calling this through the MCP. The semble process even sticks around as a zombie, forever stalled out. No idea why, logs have nothing. When called…
cityofdelusion on HN
Questions
Does it support multiple languages?
From 1 comment
Does this support any language or is it limited to a specific set of languages?
gslepak on HN
How does it compare with LSPs?
From 1 comment
Better than grep obviously, but how does this compare to existing LSPs?
AussieWog93 on HN
Security and supply-chain risk
From 1 comment
The one thing I'm a bit nervous about: security. Thoughts of supply-chain "what-ifs" gives me a bit of pause here. Would like to hear security-minded folks give their thoughts on…
handonam on HN
Alternatives named
grep and ripgrep
From 2 comments
I'd be interested in seeing actual agent benchmarks (eg CC or Copilot CLI with grep removed and this tool instead). For example, I have explored RTK and various LSP…
jerezzprime on HNFeedback: codex-cli hangs when calling this through the MCP. The semble process even sticks around as a zombie, forever stalled out. No idea why, logs have nothing. When called…
cityofdelusion on HN
codebase-memory-mcp
From 2 comments
What I have personally observed with such tools is that they make the AI's dumb, similar to how it makes coders dumb when relying more on AI tools. These agentic AI's are already…
freakynit on HNWould this replace something like codebase-memory-mcp[1] or improve when both is being used? [1] - https://github.com/DeusData/codebase-memory-mcp
_ink_ on HN
Serena
From 1 comment
Serena does a lot more than semble (I actually used serena before building this and didn't like how much it does by default). That also made it hard to see if it was actually…
Bibabomas on HN
colgrep
From 1 comment
How does this compare with colgrep? https://github.com/lightonai/next-plaid/tree/main/colgrep
nextaccountic on HN
context-mode
From 1 comment
How does it compare to context-mode or serina that are both well established now?
smcleod on HN