Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/lktiep/cortex-hub/csgit clone --depth 1 https://github.com/lktiep/cortex-hubWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00924 |
| Opus 5 | $0.00000 | $0.00462 |
| Sonnet 5 | $0.00000 | $0.00185 |
| Haiku 4.5 | $0.00000 | $0.00092 |
Grade A, and why
cs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/cs — Cortex Start v0.7.0
Version: 0.7.0 | Updated: 2026-04-11 Changelog: v0.7.0 — unified versioning, removed STATE.md, streamlined tool guidance, auto-memory safety net Changelog: v2.1 — added plan quality gate before implementation Changelog: v2.0 — added task pickup, detect changes, recipe health, workflow recipes, versioning
Run ALL steps IN ORDER. Do NOT proceed to user work until Step 7 completes.
Step 1: Session Start
Call cortex_session_start:
repo: "https://github.com/lktiep/cortex-hub.git"
mode: "development"
agentId: "claude-code"
ide: "<your IDE>"
branch: "<current git branch>"
Save session_id and projectId from the response.
If recentChanges.count > 0 → warn user and git pull before any edits.
Step 2: Recall Context (parallel)
Call BOTH in parallel:
cortex_knowledge_search(query: "session summary progress next session")cortex_memory_search(query: "session context decisions lessons", agentId: "claude-code")
These return what was done last session, key decisions, and next steps.
Step 3: Conflict Check
cortex_changes(agentId: "claude-code", projectId: "<from step 1>")
Step 4: Task Pickup (Optional)
If cortex_task_pickup tool is available:
cortex_task_pickup() — check for Conductor tasks assigned to you.
If tasks exist → list them. Ask user which to work on, or continue with their request.
If the tool is not available (e.g. in solo dev mode), skip this step.
Step 5: Working State Check
Run git status. If uncommitted changes:
cortex_detect_changes(scope: "all")— analyze risk level- Report affected symbols and blast radius
Step 6: Situational Summary
Print a concise report:
## Session Init Complete
- **Last session**: <what was done, from memory/knowledge recall>
- **Pending tasks**: <N tasks> or none
- **Unseen changes**: <from other agents> or clean
- **Working state**: clean / <N uncommitted files, risk level>
- **Key context**: <relevant decisions or lessons>
- Ready to start work.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 95 lines · 0 tokens per session scan A 4bc850db1ff1
cs is a command published in the GitHub repository lktiep/cortex-hub (58 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 924 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
orchestrator
You are Orchestrator, the autonomous operations manager for this project. You coordinate AI agent roles to execute tasks with minimal CEO involvement.
quick
Scaffold a level-0 lawbook change (record.md + reports) for a tiny fix.
draft
Draft a new spec-driven change at the confirmed ceremony level before coding.
archive
Finalize a completed change — sync specs when needed, then archive it.
build
Implement a drafted change, following its spec and the project's standards.
explore
Enter explore mode — a thinking partner before or during a change.