
Hunta Instinct
What it can never forget.
Experience becomes recall. Recall becomes instinct.
The rare, high-severity constants your agents read before every action. Not retrieved on request. Consulted every turn.
How it works
From a costly mistake to a standing reflex
A lesson earns its place by severity. Once promoted, it stops being something to recall and becomes something read before the agent acts.
The difference
Recalled when relevant, or consulted every turn
Gather returns what fits the moment. Retrieval is scored by relevance, so the right facts surface when the query calls for them. Powerful, but conditional.
Instinct does not wait to be relevant. High-severity constants are injected before every action, so the lesson applies whether or not the prompt mentions it.
What is shipped
Every entry, every turn is live today
Severity ranking and every-turn injection are shipped. The rest is on the roadmap, and we say which is which.
Live
- Severity-ranked constants
- Every-turn injection before actions
Planned
- Conflict resolution: newer lessons replace the constants they contradict
- Automatic severity scoring, inferred from signals instead of set by hand
- Read-frequency weighting
Reflex check
Read before the agent acts
# Instinct, injected before your agent acts on "deploy to prod":
[severity 5] never deploy without the staging gate green
[severity 4] rotate the signing key before cutover
read every turn, before the action. non-negotiable.Get started
Connect Instinct to your agent
Instinct is included with Hunta Gather, so one connection covers both. Pick your client, create an API key, and your agent reads its constants before every action.
Hunta Gather
Agent memory with a receipt
Hunta Gather stores what your agents learn and recalls the current truth. Connect it to any agent harness, then run the check that sets it apart: verify your tenant's data is isolated, and get back an Ed25519-signed result you can check yourself.
token to first recall in five minutes
/plugin marketplace add hunta-ai/gather/plugin install gather@huntaexport GATHER_URL=https://mcp.hunta.aiexport GATHER_TOKEN=<your-agent-key># ~/.codex/config.toml[mcp_servers.gather]url = "https://mcp.hunta.ai/mcp" [mcp_servers.gather.http_headers]Authorization = "Bearer <YOUR_KEY>"{ "mcpServers": { "gather": { "url": "https://mcp.hunta.ai/mcp", "headers": { "Authorization": "Bearer <YOUR_KEY>" } } } }# hands-free install: point your agent at hunta.ai/connect/agent{ "mcpServers": { "gather": { "url": "https://mcp.hunta.ai/mcp", "headers": { "Authorization": "Bearer <YOUR_KEY>" } } } }# gathercurl -X POST https://mcp.hunta.ai/v1/memories -H "Authorization: Bearer $GATHER_TOKEN" -d '{"text":"Our design partner is Acme Corp."}'# recallcurl -X POST https://mcp.hunta.ai/v1/memories/search -H "Authorization: Bearer $GATHER_TOKEN" -d '{"query":"design partner"}'# verify (the beat a plain memory API cannot show)curl -X POST https://mcp.hunta.ai/v1/admin/verify-isolation -H "Authorization: Bearer $GATHER_TOKEN"