[add-routines] add routines/ skeleton — scheduled work on this repo - #27
Merged
Conversation
Adds the routines/ top-level folder. Routine = trigger + skill chain + output contract. Procedural, not autonomous LLM loop. Borrowed from Knowcap's own Routine→Skill→Run→Inbox architecture (docs/decisions/2026-05-29-agent-skills-routines-architecture.md in the main repo). Why "routines" not "agents": - "agent" is overloaded — Claude Code subagents, shipped product agents, autonomous LLM agents all collide. - Hassan already operates 9 cloud routines in av-claude-workspace/ — same vocabulary. - Knowcap product's own model uses this exact word for the same pattern. Layout: - routines/README.md — index, pattern explanation - routines/_skills/ — reusable skill definitions (the WHAT) - write-blog-draft/SKILL.md - audit-seo/SKILL.md - refresh-persona/SKILL.md - routines/<slug>/ — one folder per routine (the WHEN + WHAT reference + I/O contract) - weekly-blog/ — Mon 06:00 UTC, opens blog draft PR - nightly-seo-audit/ — Daily 03:00 UTC, opens regression-report PR if signal drops - persona-refresh/ — Monthly, validates persona segmentation against fresh signal - content-curator/ — Event-triggered (push to docs/research/), suggests blog topics Each routine has ROUTINE.md (human-readable spec) + triggers.yml (cron/event config). Each skill has SKILL.md (instructions). Status: P0 — structure + stub definitions. No runtime executes them autonomously yet. To run manually today: open Claude Code in this repo, point at a routine folder, ask Claude to follow the instructions. P1 (this week): flesh out write-blog-draft + audit-seo SKILL.md with real prompts. P2 (next 2 weeks): wire one routine to Hermes or GitHub Actions cron for autonomy. .gitignore: routines/**/runs/ and routines/**/baselines/ — per-routine state not tracked. CLAUDE.md: updated to mention the new routines/ folder. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Warning Review limit reached
More reviews will be available in 2 minutes and 51 seconds. Learn how PR review limits work. Your organization has run out of usage credits. Purchase more in the billing tab. ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (14)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Smetools
added a commit
that referenced
this pull request
Jun 28, 2026
…nt, bilingual, GEO) (#94) Distilled from field playbooks + our own 16-post / 0-click evidence. The blogger stopped inventing thesis essays; it now answers the live SERP in the language the demand is in. AGENT.md + ROUTINE.md: - read ../claude-knowcap/.../seo/PLAYBOOK.md first; playbook wins on conflict - classify INTENT (nav→skip, commercial→money-page, informational→post) - pick LANGUAGE from demand (native AR post for AR demand; hreflang pair) - STEAL THE SERP (inherit the seo routine's content-brief, else pull live top-3) is now a mandatory gate — no draft from imagination - hand the published URL to the seo routine for request-indexing (Law 6) write-blog-draft SKILL.md: - new `money-page` mode (the only format ranking #1 for us) - entry gates: ICP + serp-steal + intent - serp_steal input (brief or live top-3) drives structure + word-count target - bilingual frontmatter (lang/dir/hreflang), native-Arabic voice note - thesis prompt is now SERP-grounded; gates check SERP coverage + citable passages Pairs with claude-knowcap PR #27 (the shared PLAYBOOK + seo routine). Claude-Session: https://claude.ai/code/session_01TwM4VYoHip5iUguBWCqStH Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds the
routines/top-level folder — scheduled / triggered work that operates on this repo.Routine = trigger + skill chain + output contract. Procedural, not autonomous LLM loop. Borrowed from Knowcap's own Routine→Skill→Run→Inbox architecture decision in the main repo.
Why "routines" not "agents"
av-claude-workspace/— same vocabulary, no new term to learnLayout
Each routine has
ROUTINE.md(human-readable spec) +triggers.yml(cron / event config).Each skill has
SKILL.md(instructions).Status: P0 — structure only
No runtime executes these autonomously yet. To run manually today: open Claude Code in this repo, point at a routine folder, ask Claude to follow the instructions.
write-blog-draft+audit-seoSKILL.mdwith real prompt contentType
chore— repo restructure, no functional changeTest plan
app/).gitignoreupdated to excluderoutines/**/runs/androutines/**/baselines/CLAUDE.mdupdated to mention the new folderCompanion PR (coming next)
Same skeleton in
Knowcap-V2/knowcap(main app repo) — different routine list (feature-builder, bug-triager, pr-pre-review, docs-keeper).🤖 Generated with Claude Code