bump
Skill4.8k starsChanged 14 days agoMITView on GitHub
Bump a pinned dependency version. Use harness.md to bump an agent harness (codex, claude-code, pi, hermes-agent, prime-agent, mini-swe-agent, ...).
11 other projects use a version of this.
Adopted by
Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data
Most recently taken up by
- pandazki/pneuma-skillstheir own version, first seen Nov 6, 2026
- zanellig/skillstheir own version, first seen Oct 29, 2026
- jwilger/agent-skillstheir own version, first seen Oct 11, 2026
- poco-ai/Agenterotheir own version, first seen Oct 8, 2026
- av1155/houndarrtheir own version, first seen Oct 8, 2026
- PinW/whisper-key-localtheir own version, first seen Oct 8, 2026
Who's talking about PrimeIntellect-ai/verifiers
- A Hacker News story“Show HN: I built an integration for RL training of browser agents for everyone”Hacker News · Mar 25, 2026 · 7 points
- A Hacker News story“Verifiers: Environments for LLM Reinforcement Learning”Hacker News · Sep 26, 2025 · 2 points
- PrimeIntellect-ai/prime-agent“Preserve GPT reasoning across Responses turns”GitHub · Nov 7, 2026 · 1 reactions
- PrimeIntellect-ai/prime-envs“Add Terminal-Bench main with native Prime Compose”GitHub · Oct 31, 2026 · 1 reactions
- PrimeIntellect-ai/prime-envs“Run EnterpriseOps-Gym on native Prime VMs”GitHub · Oct 19, 2026 · 1 reactions
The biggest 5 of 53 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 53
Talked about on GitHub, Hacker News and Bluesky (53 times).
On GitHub · 50
- Nov 7, 2026Preserve GPT reasoning across Responses turns — PrimeIntellect-ai/prime-agent1 reaction · 1 comment
- Oct 31, 2026Add Terminal-Bench main with native Prime Compose — PrimeIntellect-ai/prime-envs1 reaction · 2 comments
- Oct 19, 2026Run EnterpriseOps-Gym on native Prime VMs — PrimeIntellect-ai/prime-envs1 reaction · 2 comments
- May 12, 2026Pass hosted env request timeout through — PrimeIntellect-ai/prime1 reaction · 1 comment
- Apr 13, 2026release: tunnel 0.1.6 — PrimeIntellect-ai/prime1 reaction
45 more on GitHub
- Apr 13, 2026Add Tunnel.check_registered() for server-side liveness — PrimeIntellect-ai/prime1 reaction
- Feb 18, 2026pass json_logging to env servers — PrimeIntellect-ai/prime-rl1 reaction
- Nov 6, 2026fix: preserve resumed task state when dataset rows move — PrimeIntellect-ai/prime-rl0 reactions
- Nov 6, 2026feat: support You.com Eco Search as a search provider — PrimeIntellect-ai/nano-rlm0 reactions
- Nov 6, 2026Add a verifiers v1 / vf-eval wrapper for public FDA-Gym — juanpablozarza/fda-gym0 reactions
- Nov 1, 2026build(deps-dev): bump the routine-version-updates group across 1 directory with 5 updates — kimpton-ai/environment-harness0 reactions
- Oct 30, 2026chore(deps): bump verifiers to 076c364e for rlm network guidance — PrimeIntellect-ai/prime-rl0 reactions
- Oct 30, 2026feat: train on explicit multimodal placeholder positions, not image token ids — PrimeIntellect-ai/prime-rl0 reactions
- Oct 30, 2026feat: report media placeholder positions in logical prompts; drop mm_token_type_id_map — PrimeIntellect-ai/renderers0 reactions
- Oct 27, 2026fix: apply session network blocklists to search results — PrimeIntellect-ai/nano-rlm0 reactions
- Oct 21, 2026build(deps-dev): update verifiers requirement from <0.3,>=0.2.0 to >=0.2.0,<0.5 — SourceShift/mini-ork0 reactions
- Oct 20, 2026chore: bump verifiers and prime-envs — PrimeIntellect-ai/prime-rl0 reactions
- Oct 19, 2026Bump verifiers from 0.3.0 to 0.4.0 in /environments/mazebench — zaixiakongyiji/mazebench-local-mcp0 reactions
- Oct 19, 2026Bump verifiers from 0.3.0 to 0.4.0 in /environments/mazebench — mazebench/MazeBenchEngine0 reactions
- Oct 12, 2026fix(enterprise-ops-gym): forward Prime credentials to the tool server — PrimeIntellect-ai/prime-envs0 reactions · 2 comments
- Oct 8, 2026[Integration] PrimeIntellect-ai/prime-agent: keyless You.com option for skills/websearch — youdotcom-oss/integration-tracking0 reactions · 4 comments
- Oct 8, 2026fix: retain explicitly scored execution timeouts in training — PrimeIntellect-ai/prime-rl0 reactions
- Oct 6, 2026perf(client): read prime-rl's packed completion_logprobs — PrimeIntellect-ai/renderers0 reactions · 1 comment
- Oct 3, 2026[Add] verifiers (PrimeIntellect-ai/verifiers) — PyTorchKR/landscape0 reactions
- Oct 3, 2026fix(environments): skip FIFOs and device nodes when extracting transfer archives — harbor-framework/harbor0 reactions · 2 comments
- Oct 1, 2026feat(orchestrator): support MITO training with exact engine tokens — PrimeIntellect-ai/prime-rl0 reactions
- Sep 30, 2026chore(deps-dev): update verifiers requirement from <0.3,>=0.2.0 to >=0.2.0,<0.4 — SourceShift/mini-ork0 reactions · 1 comment
- Sep 28, 2026Add ProgramBench Harbor tasks and 200 public Prime verifier images — PrimeIntellect-ai/prime-tasks0 reactions
- Sep 28, 2026feat(config): per-directory team and context (.prime/context.json) — PrimeIntellect-ai/prime0 reactions · 9 comments
- Sep 27, 2026fix: use certifi roots for model client TLS — PrimeIntellect-ai/nano-rlm0 reactions
- Sep 15, 2026Add Physim experimental prediction environment — PrimeIntellect-ai/residency-environments0 reactions · 1 comment
- Sep 13, 2026chore(deps): bump the dependencies group across 1 directory with 2 updates — existential-birds/daydream0 reactions · 3 comments
- Sep 12, 2026chore(deps): bump verifiers from 0.2.1 to 0.3.1 in /rl/daydream_review in the dependencies group — existential-birds/daydream0 reactions · 1 comment
- Sep 7, 2026omp2: bounded recursive context execution and standardized RLM/swarm evaluations are unproved — gagan114662/oh-my-pi0 reactions · 3 comments
- Sep 6, 2026exp: pin completion and routing dependency stack for OpenSeeker RL — PrimeIntellect-ai/prime-rl0 reactions
- Sep 5, 2026feat: expose structural completion status from parsers — PrimeIntellect-ai/renderers0 reactions
- Sep 2, 2026docs(rfd): propose model request relationships — agentclientprotocol/agent-client-protocol0 reactions · 1 comment
- Sep 1, 2026Bump verifiers from 0.3.0 to 0.3.1 in /Evals/marginbench in the marginbench-python group — openprose/margin0 reactions
- Aug 31, 2026Bump verifiers from 0.3.0 to 0.3.1 in /environments/mazebench — zaixiakongyiji/mazebench-local-mcp0 reactions · 1 comment
- Aug 31, 2026Bump verifiers from 0.3.0 to 0.3.1 in /environments/mazebench — mazebench/MazeBenchEngine0 reactions · 1 comment
- Aug 31, 2026chore(deps): bump the dependencies group across 1 directory with 3 updates — existential-birds/daydream0 reactions · 1 comment
- Aug 20, 2026Train from inline multimodal rollouts — PrimeIntellect-ai/prime-rl0 reactions
- Aug 20, 2026Add raw multimodal content parts support — PrimeIntellect-ai/renderers0 reactions · 1 comment
- Aug 15, 2026[daydream] Isolate RL reward scoring against mutable agent state (consolidates #450 #452 #453) — existential-birds/daydream0 reactions · 8 comments
- Aug 12, 2026chore(deps-dev): update verifiers requirement from <0.3,>=0.2.0 to >=0.2.0,<0.4 — SourceShift/mini-ork0 reactions · 1 comment
- Aug 12, 2026Demote post-change suite results to versioned non-regression telemetry — existential-birds/daydream0 reactions · 3 comments
- Aug 11, 2026Github trending [PrimeIntellect-ai/prime-agent] 开源长期任务编码代理 — web1992/github-trending0 reactions
- Aug 10, 2026Github trending [PrimeIntellect-ai/prime-agent] 开源持久化编码智能体 — web1992/github-trending0 reactions
- Aug 9, 2026Github trending [PrimeIntellect-ai/prime-agent] 递归语言模型自动优化 — web1992/github-trending0 reactions
- Aug 8, 2026Github trending [PrimeIntellect-ai/prime-agent] 开源长期任务编码智能体 — web1992/github-trending0 reactions
On Hacker News · 2
- Mar 25, 2026Show HN: I built an integration for RL training of browser agents for everyone7 points · 1 comment
- Sep 26, 2025Verifiers: Environments for LLM Reinforcement Learning2 points
On Bluesky · 1
- Jul 23, 2026「環境で鍛える」。モデルに問題を大量投入する話だと思っていたけど、Verifiersを読むと少し違う。環境はデータセット、モデル用ハーネス、採点ルーブリックの組み合わせ。 https://github.com/PrimeIntellect-ai/verifiers… — @nemu-echosphere.bsky.social1 likes and repost · 1 reply
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 11
- poco-ai/Agentero★ 2ksame name
- av1155/houndarr★ 308same name
- PinW/whisper-key-local★ 203same name
- pandazki/pneuma-skills★ 161same name
- GeoloeG-IsT/agents-reverse-engineer★ 20same name
- maxwellsdm1867/wheeler★ 12same name
- atournayre/claude-marketplace★ 9same name
- soderlind/skills★ 5same name
- jwilger/agent-skills★ 4same name
- zanellig/skills★ 1same name
1 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
Files
Bump
- To bump an agent harness version, follow harness.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

