diff options
Diffstat (limited to 'gemfeed')
| -rw-r--r-- | gemfeed/2026-07-09-unveiling-hexai.gmi | 5 | ||||
| -rw-r--r-- | gemfeed/2026-07-09-unveiling-hexai.gmi.tpl | 5 | ||||
| -rw-r--r-- | gemfeed/atom.xml | 7 |
3 files changed, 16 insertions, 1 deletions
diff --git a/gemfeed/2026-07-09-unveiling-hexai.gmi b/gemfeed/2026-07-09-unveiling-hexai.gmi index f3a5e381..90d4afea 100644 --- a/gemfeed/2026-07-09-unveiling-hexai.gmi +++ b/gemfeed/2026-07-09-unveiling-hexai.gmi @@ -326,6 +326,11 @@ Urgency: 7.8 Tags: agent ``` +This is where it becomes useful for agentic coding. There is an agent skill and a slash command built on top of `ask`: the `agent-task-management` skill knows how to read and update the `ask` task list, and the `/work-on-tasks` command uses that skill to work through a whole backlog on its own — pick the next actionable task, hand it to a fresh sub-agent, mark it done, and continue until nothing actionable is left. That is pretty useful after a planning session where you created a bunch of tasks with their proper dependencies set: point the agent at the project, run `/work-on-tasks`, and it chews through the queue one task at a time. + +=> https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/skills/agent-task-management agent-task-management skill +=> https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/commands/work-on-tasks.md /work-on-tasks command + ## Running multiple providers side by side One feature I use more than I expected: per-surface model lists. In `config.toml` you can configure several providers or models for the same entry point, and Hexai fans the request out to all of them in parallel. diff --git a/gemfeed/2026-07-09-unveiling-hexai.gmi.tpl b/gemfeed/2026-07-09-unveiling-hexai.gmi.tpl index d7c34a77..27c7291e 100644 --- a/gemfeed/2026-07-09-unveiling-hexai.gmi.tpl +++ b/gemfeed/2026-07-09-unveiling-hexai.gmi.tpl @@ -312,6 +312,11 @@ Urgency: 7.8 Tags: agent ``` +This is where it becomes useful for agentic coding. There is an agent skill and a slash command built on top of `ask`: the `agent-task-management` skill knows how to read and update the `ask` task list, and the `/work-on-tasks` command uses that skill to work through a whole backlog on its own — pick the next actionable task, hand it to a fresh sub-agent, mark it done, and continue until nothing actionable is left. That is pretty useful after a planning session where you created a bunch of tasks with their proper dependencies set: point the agent at the project, run `/work-on-tasks`, and it chews through the queue one task at a time. + +=> https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/skills/agent-task-management agent-task-management skill +=> https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/commands/work-on-tasks.md /work-on-tasks command + ## Running multiple providers side by side One feature I use more than I expected: per-surface model lists. In `config.toml` you can configure several providers or models for the same entry point, and Hexai fans the request out to all of them in parallel. diff --git a/gemfeed/atom.xml b/gemfeed/atom.xml index 4d8934b2..13b1ef7e 100644 --- a/gemfeed/atom.xml +++ b/gemfeed/atom.xml @@ -1,6 +1,6 @@ <?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> - <updated>2026-07-08T17:18:58+03:00</updated> + <updated>2026-07-08T17:26:44+03:00</updated> <title>foo.zone feed</title> <subtitle>To be in the .zone!</subtitle> <link href="gemini://foo.zone/gemfeed/atom.xml" rel="self" /> @@ -373,6 +373,11 @@ Urgency: 7.8 Tags: agent </pre> <br /> +<span>This is where it becomes useful for agentic coding. There is an agent skill and a slash command built on top of <span class='inlinecode'>ask</span>: the <span class='inlinecode'>agent-task-management</span> skill knows how to read and update the <span class='inlinecode'>ask</span> task list, and the <span class='inlinecode'>/work-on-tasks</span> command uses that skill to work through a whole backlog on its own — pick the next actionable task, hand it to a fresh sub-agent, mark it done, and continue until nothing actionable is left. That is pretty useful after a planning session where you created a bunch of tasks with their proper dependencies set: point the agent at the project, run <span class='inlinecode'>/work-on-tasks</span>, and it chews through the queue one task at a time.</span><br /> +<br /> +<a class='textlink' href='https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/skills/agent-task-management'>agent-task-management skill</a><br /> +<a class='textlink' href='https://codeberg.org/snonux/dotfiles/src/branch/master/prompts/commands/work-on-tasks.md'>/work-on-tasks command</a><br /> +<br /> <h2 style='display: inline' id='running-multiple-providers-side-by-side'>Running multiple providers side by side</h2><br /> <br /> <span>One feature I use more than I expected: per-surface model lists. In <span class='inlinecode'>config.toml</span> you can configure several providers or models for the same entry point, and Hexai fans the request out to all of them in parallel.</span><br /> |