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 skills/amitvijapur/cortex/cortex-reroutenpx skills add amitvijapur/cortex --skill cortex-reroutegit clone --depth 1 https://github.com/amitvijapur/cortexWhat 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 | $0.00052 | $0.00612 |
| Opus 5 | $0.00026 | $0.00306 |
| Sonnet 5 | $0.00010 | $0.00122 |
| Haiku 4.5 | $0.00005 | $0.00061 |
Grade A, and why
cortex-reroute 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/cortex-reroute
Flag the most recent entry in ~/.claude/cortex-log.jsonl as outcome=corrected and record the route the user wanted instead. This is the highest-quality training signal for Phase 3 (similarity bias) when it activates.
When to invoke
- User types
/cortex-reroute "<corrected route>"explicitly - User says "reroute" / "actually use X" / "we should've used Y" right after a routing line
- User says "the right route was..." / "should've been..." / "redo with..."
- Any clear post-hoc correction of the last logged route
What to do
Step 1 — Mark the last entry as corrected
python3 ~/.claude/bin/cortex reroute --to "<the route the user wanted>"
The CLI:
- Sets
outcome=correctedon the last log entry - Stores the user's preferred route in
user_correction - Prints the before → after delta
Step 2 — Log the new route
If the user wants the corrected route to actually run now, log it as a fresh entry with --redirect-from so the two are linked:
python3 ~/.claude/bin/cortex log \
--task "<same task as before>" \
--class <class> \
--system <new system> \
--pattern <new pattern> \
--agent "<new agent>" \
--tier <new tier> \
--tier-reason "<...>" \
--system-reason "User redirected from <old route>" \
--redirect-from "<old route>"
The redirect_from field links the corrected entry to the new one — Phase 3 uses this to learn "Cortex picked X, user wanted Y" deltas.
Step 3 — Course-correct the work
Carry out the user's preferred route. Don't argue with the correction unless there's a safety reason — the correction signal is sacred.
Quote format for the --to argument
<System> > <Pattern> [> <Agent>] @ L<n> [optional bracketed tags]
Do NOT
- Do not invoke this for routes that ran fine — only for explicit corrections
- Do not modify older log entries; this only touches the most recent
- Do not skip step 2 if the user wants the corrected route to run — without
--redirect-from, the link between old and new is lost - Do not push back on the correction — the user's choice wins
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 · 66 lines · 52 tokens per session scan A e05f7c31a03c
cortex-reroute is a skill published in the GitHub repository amitvijapur/cortex (5 stars, last pushed 2d ago), licensed MIT. It adds 52 tokens to every session and 612 once invoked, about $0.0003 per session on Opus 5. 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.