Commit Graph

14 Commits

Author SHA1 Message Date
rcourtman f6a289f72d Make Patrol report confirmed health failures 2026-07-14 18:04:09 +01:00
rcourtman 8c8affc041 Split tool registration authority and make the registry append-only
Review of 3073a5061 found the remaining registration hole: Register
rejected canonical descriptor overrides but still accepted a canonical
NAME with a nil override, inheriting the canonical descriptor while
replacing the governed handler in the map - an extension could re-register
pulse_read and bypass its execution-intent enforcement.

Registration authority is now split. registerBuiltin is the unexported
construction-time path for canonical Pulse tools: shared descriptor
mandatory, overrides rejected. RegisterExtension - the only path exposed
through PulseToolExecutor.RegisterTool - rejects every canonical tool
name outright and requires the extension to declare its own descriptor.
Both paths are append-only: a name registers exactly once, so no later
registration can swap out an already-governed handler.

Proofs cover the exact bypass (canonical name, nil override), extension
and builtin duplicate rejection, builtin override rejection, and
descriptor-less extension rejection. Tests that previously swapped
handlers by re-registering now use fresh executors per scenario.
Contract prose and source pins updated.
2026-07-10 15:16:06 +01:00
rcourtman ee8a24e14a backend and governance: MCP contract, agent capabilities, API, and release-control
Manifest-backed MCP tools, prompts, and resources with surface affordance contracts; agent capability manifest and governance projection; API contract tests and capability route projection; operations-loop and intelligence-funnel telemetry; release-control subsystem documentation, registry, and tooling; licensing and configuration.
2026-06-23 17:26:15 +01:00
rcourtman 29ac5945bc Fix resource-context Assistant tool targeting 2026-06-04 19:18:08 +01:00
rcourtman 0561a1b5fa Remove Assistant and Patrol tool-routing heuristics 2026-05-15 14:51:15 +01:00
rcourtman 895b78bd89 Make Assistant tool choice model-owned 2026-05-15 11:27:15 +01:00
rcourtman f8fd641978 Harden Pulse Assistant action governance 2026-04-23 23:01:51 +01:00
rcourtman a41c956320 fix(ai): fail closed on unknown v6 read commands 2026-04-09 19:21:16 +01:00
rcourtman 21fa343fa1 Enable structured AI auto-recovery paths 2026-03-31 09:24:56 +01:00
rcourtman 046a0e92c0 Normalize pulse_read native log denial paths 2026-03-31 09:07:57 +01:00
rcourtman 298b23626b feat(ai): add canonical truenas app log reads 2026-03-29 20:13:39 +01:00
rcourtman 778a2577b6 feat: Pulse v6 release 2026-03-18 16:06:30 +00:00
rcourtman 95a0d7a6bd feat(backend): implement AI Patrol, Investigation, and system-wide refactors 2026-01-30 19:02:14 +00:00
rcourtman b2e0ae3fdb Add ExecutionIntent classification and NonInteractiveOnly enforcement
Implement safety layers for command execution:

ExecutionIntent classifies commands as:
- ObservationOnly: Pure read (status, logs, metrics)
- SideEffects: May change state (restart, write, delete)

NonInteractiveOnly enforces safe command forms:
- Blocks interactive commands (vim, top without -b, etc)
- Blocks unbounded streaming (tail -f without limit)
- Suggests safe alternatives in error messages

Add phantom execution detection:
- Catches when model claims actions without using tools
- Skips check when tools actually succeeded (fixes false positives)

Includes comprehensive tests for:
- Intent classification accuracy
- Interactive command blocking
- Strict resolution validation
2026-01-28 16:49:00 +00:00