Peter Dave Hello͌ͯͦ̉

4K posts

Peter Dave Hello͌ͯͦ̉ banner
Peter Dave Hello͌ͯͦ̉

Peter Dave Hello͌ͯͦ̉

@PeterDaveHello

Free and open source software enthusiast, Maintainer and active contributor of few projects.

Taiwan, R.O.C Katılım Aralık 2014
2.7K Takip Edilen709 Takipçiler
Tibo
Tibo@thsottiaux·
What’s something we haven’t fixed in codex for a while and that’s plain annoying?
English
2.3K
34
2K
337.9K
nader dabit
nader dabit@dabit3·
Claude Opus 4.8 is now available in Devin CLI ⚡️ I'm giving away 10 @devinai Max subscriptions ($200 each) for you to try it out! Just comment below with what you're building or want to build to enter!
nader dabit tweet media
English
703
47
976
83K
Pullfrog
Pullfrog@pullfrogai·
introducing Pullfrog: the OSS alternative to CodeRabbit that lives in GitHub Actions 🐸 ⚖️ model-agnostic 💰 7¢/run, BYOK or at-cost model router ⚡ sub-30s reviews + one-click Fix; subagent swarms for big PRs 🤖 tag @pullfrog for issue triage, CI auto-fix, custom crons 🔋 batteries included: MCP tools, headless browser, repo memory 🪝 lifecycle hooks for linting, testing, and scripts 🔒 isolated shell, run-scoped tokens, branch protections 🥷 secret auto-masking — zero creds exposed to the agent 👥 built on GH perms — no redundant seats/roles 📊 org/repo analytics + monthly billing quotas
Pullfrog tweet media
English
6
4
71
18.6K
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
請讓 Codex 全面檢視你的對談紀錄(sessions)、記憶(Memories)與歷程紀錄(Chronicle),找出反覆出現的模式,盡可能沿用既有內容,並且只建立規模最小、但確實有用的技能(skill)、子代理人(subagent)或自動化流程(automation)。 「請回顧我過去 30 天內的近期工作;如果可用紀錄不足 30 天,就檢視所有可用紀錄。請找出值得封裝成可重複使用流程的手動工作。 請依照以下順序參考可用依據: 1. 近期的 Codex 對談紀錄與任務摘要。 2. Codex 記憶(Memories)與 rollout 摘要,用來找出跨對談反覆出現的模式。 3. 歷程紀錄(Chronicle);若已啟用,請用它發掘 Codex 以外的重複工作。Chronicle 僅作為發現線索之用;重要細節請盡可能回到相關的原始來源系統確認。 4. 既有的技能、自訂代理人與自動化流程,以便優先沿用或擴充現有內容,避免重複造輪子。 請廣泛尋找符合以下特徵的工作:重複發生、耗時、容易出錯、高度仰賴背景脈絡,或明顯能從一致流程中受益。範圍包含寫程式、研究、寫作、規劃、溝通、營運、分析,以及個人行政庶務等工作流程。 只有在候選工作符合以下條件時,才採取行動: - 至少出現過兩次,或明顯很可能再次發生,而且每次重做成本高; - 輸入條件穩定、處理步驟可重複,並且有明確的產出或停止條件; - 能實質提升處理速度、品質、一致性或可靠性; - 既有內容尚未充分涵蓋它。 請選擇規模最小且最合適的形式: - 技能(Skill):可重複使用的工作流程或操作手冊(playbook)。 - 自訂子代理人(Custom subagent):邊界明確、適合交辦的專責角色或調查任務。 - 自動化流程(Automation):排程或定期執行的檢查、報告、提醒或監控。 - 跳過(Skip):太過一次性、模糊、敏感,或佐證不足,因而不適合封裝的工作。 請先產出一份精簡短名單,列出: - 重複出現的工作流程 - 佐證依據與日期 - 出現頻率/信心度 - 建議處理方式:技能、子代理人、自動化流程、擴充既有內容,或跳過 - 為什麼值得建立,或為什麼不值得建立 接著,只建立那些信心度高、且目前確實缺少的內容。請讓它們範圍限縮、實用、能辨明資料來源,且容易驗證。不要建立推測性過強、彼此重疊或範圍過大的資產。 最後請整理: - 你建立或擴充了什麼 - 你刻意跳過了什麼 - 哪些事項還需要更多佐證,才適合封裝」
Vaibhav (VB) Srivastav@reach_vb

UPDATE: Came up with an even better version of this prompt after the feedback Ask Codex to look across your sessions, Memories, and Chronicle, identify patterns, reuse what already exists, and only create the smallest useful skill, subagent, or automation. "Look back over my recent work from the last 30 days, or all available history if shorter, and identify repeated manual workflows worth packaging. Use available evidence in this order: - Recent Codex sessions and task summaries. - Codex Memories and rollout summaries to find patterns repeated across sessions. - Chronicle, if enabled, to spot repeated work outside Codex. Use Chronicle for discovery only; confirm important details in the relevant source system when possible. - Existing skills, custom agents, and automations, so you reuse or extend what already exists instead of duplicating it. Look broadly for work that is repeated, time-consuming, error-prone, context-heavy, or benefits from a consistent process. Include workflows across coding, research, writing, planning, communication, operations, analysis, and personal administration. Only act on a candidate when it: - occurred at least twice, or is clearly likely to recur and costly to repeat; - has stable inputs, a repeatable procedure, and a clear output or stopping condition; - would materially improve speed, quality, consistency, or reliability; - is not already adequately covered. Choose the smallest appropriate form: - Skill: a reusable workflow or playbook. - Custom subagent: a bounded specialist role or investigation task suitable for delegation. - Automation: a scheduled or recurring check, report, reminder, or monitor. - Skip: work that is too one-off, ambiguous, sensitive, or poorly evidenced to package. First produce a compact shortlist with: - repeated workflow - supporting evidence and dates - frequency/confidence - recommended form: skill, subagent, automation, extend existing, or skip - why it is or is not worth creating Then create only the high-confidence missing items. Keep them narrow, practical, source-aware, and easy to validate. Do not create speculative, overlapping, or overly broad assets. Finish with: - what you created or extended - what you deliberately skipped - what needs more evidence before packaging"

中文
0
0
0
148
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
@thsottiaux Please fix the randomly and intermittently occurring "bwrap: loopback: Failed RTM_NEWADDR: Operation not permitted" error in Codex 🙏
English
0
0
0
24
Tibo
Tibo@thsottiaux·
A little secret. About 5% of our production traffic is on the Pi harness, about another 5% is on OpenCode. Reminder you can use your ChatGPT account in a flourishing set of other tools. We’ll continue to make Codex awesome, but you have options.
English
406
300
8.5K
887.6K
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
@thsottiaux Codex really needs to learn when to use apps or mcps, got no idea why a local code review should search so much on GitHub 🧐
English
0
0
0
422
GitHub Changelog
GitHub Changelog@GHchangelog·
Model options for Copilot Chat on the web have been reduced to improve response consistency and quality. • Removed all Gemini models and select other models like GPT-5.2 Codex and GPT-5.4 nano github.blog/changelog/2026…
English
5
7
68
23.8K
Tibo
Tibo@thsottiaux·
Codex will do wonderful things for you if you ask nicely
English
448
60
2.3K
139.7K
Naor Peled
Naor Peled@thehecticbyte·
Some news: I'm stepping in to lead the maintenance of PR Agent 🚀 The project is moving to a community-owned org (The-PR-Agent), license going back to Apache 2.0, governance committee forming. Huge props to @QodoAI and @DanaFine7 for doing this the right way. github.com/The-PR-Agent
Naor Peled tweet media
English
2
0
3
328
GitHub
GitHub@github·
GitHub Pages is the easiest way to host your static sites for free. If you’ve already shipped yours, drop the link in the replies👇and show us what you built! 🌍
English
247
110
1.9K
218.8K
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
Sometimes, even "gpt-5.4 medium" with the "fast" mode is slow, and now the `.codex` is created everywhere... OMG...
English
0
0
0
63
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
Something's been off with the last few Codex CLI versions. Running /review in a clean session keeps triggering codex_apps.github_fetch_pr_patch or .github_search for no reason, sometimes even returning feedback completely unrelated to local changes. Totally broken and frustrating. A great feature somehow went downhill 😓
Peter Dave Hello͌ͯͦ̉ tweet media
English
3
0
1
107
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
Congrats on the 2.9 release! It's great to see the continued emphasis on performance and quality improvements. That said, the default concurrency behavior still feels a bit too aggressive. Builds can easily overwhelm system resources before there's time to intervene, which can be frustrating. A more conservative default would make a meaningful difference.
English
0
0
0
130
Turborepo
Turborepo@turborepo·
Turborepo 2.9 focuses on quality: • Up to 96% faster Time to First Task • 𝚝𝚞𝚛𝚋𝚘 𝚚𝚞𝚎𝚛𝚢 is now stable • Future Flags and deprecations ahead of 3.0 • Closes 141 Issues (70% total reduction), addressing 305 👍 reactions turborepo.dev/blog/2-9
English
10
14
348
81.6K
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
@thsottiaux I also noticed that Codex often fails to recognize which commands should prompt for privilege escalation before they run. Even when I clearly told it that a command (such as docker) involves network operations, the same frustrating issue keeps happening.
English
0
0
0
480
Tibo
Tibo@thsottiaux·
Always fun when you notice Codex being clever in a way you don't expect. In a session today, it was running a slow build process and got annoyed (don't we all). Before making a change it checked that progress was actually happening and did so not by checking the logs, but by checking CPU usage.
Tibo tweet media
English
103
19
959
66.2K
Peter Dave Hello͌ͯͦ̉
Peter Dave Hello͌ͯͦ̉@PeterDaveHello·
But regardless of which environment, I keep running into various bwrap issues in recent versions, such as `bwrap: loopback: Failed RTM_NEWADDR` or `bwrap: Unknown option --argv0`. What happened to Codex's quality control...
English
0
0
0
53