Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

AI coding agents re-explore the same codebase every single session. Repowise indexes a repo once and gives Claude Code, Codex, or any MCP agent a real dependency graph, git history, and a bug-predictive code health score. No re-grepping the same files. No stale docs. No guessing which file is... show more

Simplifying AI

37,544 subscribers

17,561 Aufrufe • vor 3 Monaten •via X (Twitter)

Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

ANTHROPIC JUST TURNED AI AGENTS INTO GIT REPOS Anthropic shipped "ant" - a CLI that runs every Claude API endpoint straight from your terminal. The headline isn't the terminal access. It's that you can now version-control an AI agent as YAML in Git and have CI sync it to the Claude Platform, the same way you ship code. - Every API resource is a subcommand: messages, models, files, agents, sessions - Define an agent in a YAML file, check it into your repo, and keep it in sync with one update command - Spin up a session, send it an event, then pull every event and tool call back from the same CLI - Claude Code knows how to drive ant out of the box - it shells out and reads the results with no glue code Agents just stopped being prompts you babysit and became infrastructure you deploy.

ANTHROPIC JUST TURNED AI AGENTS INTO GIT REPOS Anthropic shipped "ant" - a CLI that runs every Claude API endpoint straight from your terminal. The headline isn't the terminal access. It's that you can now version-control an AI agent as YAML in Git and have CI sync it to the Claude Platform, the same way you ship code. - Every API resource is a subcommand: messages, models, files, agents, sessions - Define an agent in a YAML file, check it into your repo, and keep it in sync with one update command - Spin up a session, send it an event, then pull every event and tool call back from the same CLI - Claude Code knows how to drive ant out of the box - it shells out and reads the results with no glue code Agents just stopped being prompts you babysit and became infrastructure you deploy.

BuBBliK

200,690 Aufrufe • vor 5 Monaten

if you use claude code, this will save you real money. the problem: every time you make an edit, your ai rereads the whole codebase to figure out what changed. tens of thousands of tokens, every turn, for context it already had. this repo fixes it. it’s called code review graph, and it just maps your entire codebase: every file, every function, every connection laid out so you can see the actual shape of your project how to set up (2 min): 1. pip install code-review-graph 2. code-review-graph install - auto-configures Claude Code, Cursor, Codex, Gemini CLI + more 3. code-review-graph build change one function, and it traces exactly what that touches... every caller, every dependent file, every test. your ai only reads what's affected, not the whole repo. and the savings are wild. a task that used to burn ~100,000 tokens (about a dollar) now runs closer to a penny.

if you use claude code, this will save you real money. the problem: every time you make an edit, your ai rereads the whole codebase to figure out what changed. tens of thousands of tokens, every turn, for context it already had. this repo fixes it. it’s called code review graph, and it just maps your entire codebase: every file, every function, every connection laid out so you can see the actual shape of your project how to set up (2 min): 1. pip install code-review-graph 2. code-review-graph install - auto-configures Claude Code, Cursor, Codex, Gemini CLI + more 3. code-review-graph build change one function, and it traces exactly what that touches... every caller, every dependent file, every test. your ai only reads what's affected, not the whole repo. and the savings are wild. a task that used to burn ~100,000 tokens (about a dollar) now runs closer to a penny.

Alvaro Cintas

75,596 Aufrufe • vor 3 Monaten

THIS IS WHAT A SUPPORT TEAM LOOKS LIKE IN 2026 A customer reports that their export stops at 10,000 of 18,400 contacts. And instead of passing the ticket from tool to tool: → Support agent replies → Coding agent finds the bug → Docs agent updates the guide → CRM agent prepares the renewal All on the same ticket, with the same context: logs, billing, code and previous conversations. It's called Helpin. It's open source and you can self-host it.

THIS IS WHAT A SUPPORT TEAM LOOKS LIKE IN 2026 A customer reports that their export stops at 10,000 of 18,400 contacts. And instead of passing the ticket from tool to tool: → Support agent replies → Coding agent finds the bug → Docs agent updates the guide → CRM agent prepares the renewal All on the same ticket, with the same context: logs, billing, code and previous conversations. It's called Helpin. It's open source and you can self-host it.

marcus

22,919 Aufrufe • vor 1 Monat

70% fewer tokens per task. Context that compounds across every session, every agent, every teammate. Agent traces are the memory. TOKENMAXXING for coding agents on OriginTrail DKG v10.

70% fewer tokens per task. Context that compounds across every session, every agent, every teammate. Agent traces are the memory. TOKENMAXXING for coding agents on OriginTrail DKG v10.

OriginTrail

1,360,087 Aufrufe • vor 5 Monaten

i don't f*cking understand why this isn't popular yet someone has created a memory system that uses 90% fewer tokens while still finding all the expected symbols it builds a local graph of source symbols and their relationships. architecture, decisions and handoffs live as Git-tracked Markdown. a note can point to the code behind it, so a code change can flag that knowledge for review the loop: scan the repo → map the code → write the why → anchor notes to symbols → retrieve a task-sized slice → review drift → commit to Git → continue in the next session in the author's small, one-repo test, graph retrieval returned 10.74× less context than grep top-3 while finding every expected symbol across six tasks the useful part is the connection between memory and evidence. the repo carries what the agent learned, and the code gives you a way to check whether that knowledge still holds save this, then build a second brain for your company⭣

i don't f*cking understand why this isn't popular yet someone has created a memory system that uses 90% fewer tokens while still finding all the expected symbols it builds a local graph of source symbols and their relationships. architecture, decisions and handoffs live as Git-tracked Markdown. a note can point to the code behind it, so a code change can flag that knowledge for review the loop: scan the repo → map the code → write the why → anchor notes to symbols → retrieve a task-sized slice → review drift → commit to Git → continue in the next session in the author's small, one-repo test, graph retrieval returned 10.74× less context than grep top-3 while finding every expected symbol across six tasks the useful part is the connection between memory and evidence. the repo carries what the agent learned, and the code gives you a way to check whether that knowledge still holds save this, then build a second brain for your company⭣

beamnxw ./

33,354 Aufrufe • vor 1 Monat

EVERYONE'S TRYING TO SOLVE AI TEAM MEMORY WITH SERVERS, VECTOR DATABASES, AND ORCHESTRATION PLATFORMS. THIS OPEN SOURCE TOOL DOES IT WITH ONE FOLDER IN YOUR REPO. Every dev on your team runs Claude Code. When one agent screws something up, the rest have no idea. They just repeat the mistake next week. It's called teamlore. When your agent gets corrected or breaks something, it writes a small lore file into a .lore/ folder. That file ships with your PR, gets reviewed like normal code, and after merge every teammate's agent automatically recalls it when they touch that part of the repo. No server. No datab No accounts. No SaaS bill. Just a folder in git. Which means code review catches bad lessons before they poison the team, git blame tells you when a rule was added and why, and the whole thing works offline. One command to install: npx teamlore init Companion command: npx teamlore scarmap. Turns your team's history of mistakes into a visual heat map of the codebase. Every red zone is a place your team has been burned before. Which means every red zone is a place your agents should slow down. Here's the wildest part. The teamlore repo's own .lore/ folder contains every mistake Claude made while building teamlore itself. Dogfooded end-to-end. You can literally open the folder and read the receipts. The author's public invitation: "Would love for someone to try and break it." Available on npm. Repo just launched. 100% open source. (link in the comments)

EVERYONE'S TRYING TO SOLVE AI TEAM MEMORY WITH SERVERS, VECTOR DATABASES, AND ORCHESTRATION PLATFORMS. THIS OPEN SOURCE TOOL DOES IT WITH ONE FOLDER IN YOUR REPO. Every dev on your team runs Claude Code. When one agent screws something up, the rest have no idea. They just repeat the mistake next week. It's called teamlore. When your agent gets corrected or breaks something, it writes a small lore file into a .lore/ folder. That file ships with your PR, gets reviewed like normal code, and after merge every teammate's agent automatically recalls it when they touch that part of the repo. No server. No datab No accounts. No SaaS bill. Just a folder in git. Which means code review catches bad lessons before they poison the team, git blame tells you when a rule was added and why, and the whole thing works offline. One command to install: npx teamlore init Companion command: npx teamlore scarmap. Turns your team's history of mistakes into a visual heat map of the codebase. Every red zone is a place your team has been burned before. Which means every red zone is a place your agents should slow down. Here's the wildest part. The teamlore repo's own .lore/ folder contains every mistake Claude made while building teamlore itself. Dogfooded end-to-end. You can literally open the folder and read the receipts. The author's public invitation: "Would love for someone to try and break it." Available on npm. Repo just launched. 100% open source. (link in the comments)

Harman

35,140 Aufrufe • vor 3 Monaten

you don't need to re-explain your codebase's architecture to your agent every session. most tools stop at telling you what broke. sentrux is a real-time architectural sensor, it watches your codebase as a live treemap and turns file structure and dependencies into one continuous quality score. the loop is simple: codebase > agent scans structure and dependencies > sentrux scores 5 root cause metrics into one signal > agent sees exactly where risk concentrates > next session starts from a live map instead of a blind grep the binary carries zero built-in language knowledge, all 52 languages live in plugin.toml and tags.scm query files, so a new language needs zero rust code. small catch: it only scores the structure, it won't tell you why the cycle happened, that part's still on you. built pure Rust with no runtime dependencies, specifically so it could sit as one binary between an agent and a codebase without adding friction.

you don't need to re-explain your codebase's architecture to your agent every session. most tools stop at telling you what broke. sentrux is a real-time architectural sensor, it watches your codebase as a live treemap and turns file structure and dependencies into one continuous quality score. the loop is simple: codebase > agent scans structure and dependencies > sentrux scores 5 root cause metrics into one signal > agent sees exactly where risk concentrates > next session starts from a live map instead of a blind grep the binary carries zero built-in language knowledge, all 52 languages live in plugin.toml and tags.scm query files, so a new language needs zero rust code. small catch: it only scores the structure, it won't tell you why the cycle happened, that part's still on you. built pure Rust with no runtime dependencies, specifically so it could sit as one binary between an agent and a codebase without adding friction.

Simplifying AI

18,308 Aufrufe • vor 3 Monaten

in the next version of OpenCode, file search is powered by fff - files the agent opens start ranking higher - tool calls reuse the same search layer instead of starting cold - less wasted context - fast af here’s me searching around the linux repo, a ~40m line codebase

in the next version of OpenCode, file search is powered by fff - files the agent opens start ranking higher - tool calls reuse the same search layer instead of starting cold - less wasted context - fast af here’s me searching around the linux repo, a ~40m line codebase

nexxel

254,364 Aufrufe • vor 5 Monaten

It is becoming f*cking dangerous: the NSA's own reverse-engineering tool now plugs straight into Claude Code - Ghidra has 94k stars, and GhidraMCP lets your agent read a compiled program instead of guessing what it does Install Ghidra 12.1.4 from ghidra-sre,org (free, released by the NSA) Download the GhidraMCP release and add it in File -> Install Extensions Restart Ghidra and switch on GhidraMCPPlugin in File -> Configure -> Developer Connect it: claude mcp add ghidra -- uvx --from git+ ghidra-mcp Open a binary you're allowed to analyze and let Ghidra run its analysis Ask Claude what a function does: it decompiles it, renames the variables and explains it in plain English Ask for the map: every function, import, string and cross-reference, and where the interesting code lives The decompiling runs in Ghidra on your machine, Claude just drives it Malware samples, old firmware, your own builds: your agent reads the program itself, not the docs about it Find the repo in the reply below

It is becoming f*cking dangerous: the NSA's own reverse-engineering tool now plugs straight into Claude Code - Ghidra has 94k stars, and GhidraMCP lets your agent read a compiled program instead of guessing what it does Install Ghidra 12.1.4 from ghidra-sre,org (free, released by the NSA) Download the GhidraMCP release and add it in File -> Install Extensions Restart Ghidra and switch on GhidraMCPPlugin in File -> Configure -> Developer Connect it: claude mcp add ghidra -- uvx --from git+ ghidra-mcp Open a binary you're allowed to analyze and let Ghidra run its analysis Ask Claude what a function does: it decompiles it, renames the variables and explains it in plain English Ask for the map: every function, import, string and cross-reference, and where the interesting code lives The decompiling runs in Ghidra on your machine, Claude just drives it Malware samples, old firmware, your own builds: your agent reads the program itself, not the docs about it Find the repo in the reply below

Archive

16,420 Aufrufe • vor 21 Tagen

We built the best codebase search to give any agent grounded information from GitHub repos. Introducing Sandbox Search. Point it at any repo, and we’ll spin up a secure coding agent in its own sandbox to do research for you. Use inside claude code, openclaw, cursor, and more.

We built the best codebase search to give any agent grounded information from GitHub repos. Introducing Sandbox Search. Point it at any repo, and we’ll spin up a secure coding agent in its own sandbox to do research for you. Use inside claude code, openclaw, cursor, and more.

Arlan

12,412 Aufrufe • vor 7 Monaten

175,000 $COTI is on the table. Build a private AI agent or agentic app and enter the Vibe Code Challenge: Agent Edition. No coding required. On-chain privacy tools for agents Agent Skills, MCP, and Messaging SDK Works with Claude, OpenAI, Hermes, Manus Enter now 👇

175,000 $COTI is on the table. Build a private AI agent or agentic app and enter the Vibe Code Challenge: Agent Edition. No coding required. On-chain privacy tools for agents Agent Skills, MCP, and Messaging SDK Works with Claude, OpenAI, Hermes, Manus Enter now 👇

COTI Foundation

607,372 Aufrufe • vor 4 Monaten

6. Explain code Give a link to Bard and ask it a question about a file. It'll be able to explain what it's for and how it works. Prompt → Can you explain what is the file agent py in this repo? (your link)

6. Explain code Give a link to Bard and ask it a question about a file. It'll be able to explain what it's for and how it works. Prompt → Can you explain what is the file agent py in this repo? (your link)

Paul Couvert

749,101 Aufrufe • vor 3 Jahren

Just pushed a cool update to Readout: session replays. Pick any past Claude Code session and scrub through the full timeline. Every prompt, tool call, file change. Files light up as edits land. Play back at different speeds or step through manually. →

Just pushed a cool update to Readout: session replays. Pick any past Claude Code session and scrub through the full timeline. Every prompt, tool call, file change. Files light up as edits land. Play back at different speeds or step through manually. →

Benji Taylor

378,921 Aufrufe • vor 8 Monaten

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. I wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. all of this is done using Comet's open-source project Opik. The article is quoted below.

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. I wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. all of this is done using Comet's open-source project Opik. The article is quoted below.

Akshay 🚀

50,440 Aufrufe • vor 6 Tagen

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.

Stanford researchers did it again. They just built the agent-native version of Git. When an agent works on a longer task, the run builds up a lot of state. This includes files edited/created, a dev server, a database, installed packages, KV cache, etc. Say the agent is at step 10 and makes a mistake, maybe it misreads a traceback and rewrites a file that was actually fine. The tests start failing, and the run goes off track, although everything through step eight was correct. By default, the agent just tries to fix it, which creates more edits and tool calls. This burns more tokens and grows the context. The other options are a person stepping in to redirect it or restarting the whole run from step one. That's wasteful, because it pays for every model/tool call again and re-prefills the context. Moreover, since an agent's run is non-deterministic, it doesn't reproduce the same early steps anyway. The reason it's hard to just jump back exactly to a previous correct step and resume from there is that the trajectory is only a message log. It records what the agent said and which tools it called, but not the live state underneath. That state includes things like memory, open file handles, child processes, installed packages, /tmp, and KV cache. None of that is in the log. Git can version the files, but it doesn't snapshot the running process or the KV cache. Checking out step eight moves the files back, but the process is still sitting in step-ten memory with a cold cache. Shepherd is a runtime layer by Stanford that records the run as a trace of typed events rather than a flat log. Each agent-environment interaction becomes a commit, similar to Git, but it tracks the live run. Its commit includes the agent process and the filesystem together, copy-on-write, so a branch carries the actual state and not just the files. Going back to a previous step is then a single call that forks from that commit and continues from the exact state. The copy-on-write fork is roughly five times faster than docker commit, and because the prompt prefix through step eight is unchanged, the KV cache is reused over 95% on replay, so early steps aren't reprocessed again. Once the run can be forked, a meta-agent can sit on top and operate it. It watches the trace and reverts as soon as it looks wrong, before the bad write is committed. In practice, it's just Python calling fork, replay, and revert on the trace, rather than a separate control plane wired into the harness. Not everything is reversible though. Files and sandbox changes undo themselves, but a database write has no automatic undo, so it needs a matching undo step set up in advance. Something external, like a sent email or a real charge, can't be undone, so the supervisor's job there is to catch it before it fires. They tested this on a few public benchmarks. On CooperBench, where two agents work on the same codebase, adding a live supervisor took the pair-coding pass rate from 28.8% to 54.7%. It's still early and labeled alpha. The benefit mostly shows up when a run gets branched a lot over a heavy sandbox state, which is exactly where restarting wastes the most tokens and time. If Git was made to make file changes reversible, Shepherd is trying to do the same thing for a live agent run. Shepherd Repo: (don't forget to star it ⭐ ) That said, Shepherd reverts a bad step inside a run. The harness around it, the prompts, tools, and checks the supervisor relies on, still drifts across runs as models and dependencies change. Akshay wrote about making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it can't recur. Read it below.

Avi Chawla

442,741 Aufrufe • vor 4 Monaten

GPT-5.5 in Codex is built for long-session agent work. But context still doesn't follow you across projects, agents, or teammates. ByteRover does that + Save you tokens -> No rewriting context when you switch from Claude Code to Codex. Now connecting Codex, Claude Code, OpenClaw, Hermes, and 22+ more agents. Your context moves with you. → personal context for solo work → team context for shared decisions and bug history

GPT-5.5 in Codex is built for long-session agent work. But context still doesn't follow you across projects, agents, or teammates. ByteRover does that + Save you tokens -> No rewriting context when you switch from Claude Code to Codex. Now connecting Codex, Claude Code, OpenClaw, Hermes, and 22+ more agents. Your context moves with you. → personal context for solo work → team context for shared decisions and bug history

andy nguyen

12,437 Aufrufe • vor 6 Monaten

Polymarket Agents repo is cheating code for real life If you've been thinking AI agents are cool, but what do they actually do? this Polymarket repo is the clearest answer I've seen This Polymarket repo basically hands you a mini AI hedge fund where Claude calls the shots. It uses RAG to read the news and snipe mispriced odds 24/7, handling all the messy API work so you don't have to. Your agent could literally be catching alpha while you sleep - don't miss out on this. How you'd use it in plain English: > Pick a market (or let the agent scan for opportunities). > Feed the agent context (news, social chatter, historical notes) and let the LLM decide what matters. > Turn that into a concrete trade decision and execution loop - automated Bookmark post so you don't lose the alpha

Polymarket Agents repo is cheating code for real life If you've been thinking AI agents are cool, but what do they actually do? this Polymarket repo is the clearest answer I've seen This Polymarket repo basically hands you a mini AI hedge fund where Claude calls the shots. It uses RAG to read the news and snipe mispriced odds 24/7, handling all the messy API work so you don't have to. Your agent could literally be catching alpha while you sleep - don't miss out on this. How you'd use it in plain English: > Pick a market (or let the agent scan for opportunities). > Feed the agent context (news, social chatter, historical notes) and let the LLM decide what matters. > Turn that into a concrete trade decision and execution loop - automated Bookmark post so you don't lose the alpha

BuBBliK

16,887 Aufrufe • vor 9 Monaten

Introducing Expect Let agents test your code in a real browser 1. Run Claude Code / Codex to QA your app 2. Watch a video of every bug found 3. Fix and repeat until passing Run as a CLI or agent skill. Fully open source

Introducing Expect Let agents test your code in a real browser 1. Run Claude Code / Codex to QA your app 2. Watch a video of every bug found 3. Fix and repeat until passing Run as a CLI or agent skill. Fully open source

Aiden Bai

847,772 Aufrufe • vor 7 Monaten

how I make videos with Claude in 5 steps most AI videos look the same because people skip straight to “make me a video” I run every video through the same 5 steps instead. each one hands a real file to the next > reference spec.md I show Claude a video I actually like it breaks down why it works: the pacing, where the first motion lands, the cut rhythm > data data.json real numbers before anything moves every figure has a source, so nothing on screen is made up > story story.md one point, one hero number, three beats decided before the motion, so the video is about something > motion Scene.tsx every frame written as code with Remotion and a motion-design skill nothing is random, so every word, number and timing stays editable > direct timing.ts I give notes in plain words “title is a beat late” one line changes, nothing else breaks ship final.mp4 it renders stills and checks them against the reference and the data nothing goes out until I approve it why it works: no video model, so no re-rolls same input, same result, every time you always know which step is the problem most people jump to step 4. the real difference is step 1 and step 5 full setup, prompt and the 10 open-source skills are in the article

how I make videos with Claude in 5 steps most AI videos look the same because people skip straight to “make me a video” I run every video through the same 5 steps instead. each one hands a real file to the next > reference spec.md I show Claude a video I actually like it breaks down why it works: the pacing, where the first motion lands, the cut rhythm > data data.json real numbers before anything moves every figure has a source, so nothing on screen is made up > story story.md one point, one hero number, three beats decided before the motion, so the video is about something > motion Scene.tsx every frame written as code with Remotion and a motion-design skill nothing is random, so every word, number and timing stays editable > direct timing.ts I give notes in plain words “title is a beat late” one line changes, nothing else breaks ship final.mp4 it renders stills and checks them against the reference and the data nothing goes out until I approve it why it works: no video model, so no re-rolls same input, same result, every time you always know which step is the problem most people jump to step 4. the real difference is step 1 and step 5 full setup, prompt and the 10 open-source skills are in the article

Francesco

53,708 Aufrufe • vor 29 Tagen

No coding. No data leaks. Just your agent, live on-chain. Build a private AI agent or app in the Vibe Code Challenge: Agent Edition. 175,000 $COTI in prizes Agent Skills, MCP, and Messaging SDK Works with Claude, OpenAI, Hermes, Manus Build it to win it. Enter now!

No coding. No data leaks. Just your agent, live on-chain. Build a private AI agent or app in the Vibe Code Challenge: Agent Edition. 175,000 $COTI in prizes Agent Skills, MCP, and Messaging SDK Works with Claude, OpenAI, Hermes, Manus Build it to win it. Enter now!

COTI Foundation

3,182,647 Aufrufe • vor 4 Monaten