feat(guardrails): steering channel + shell-in-file rule + capability-gap detector

Bundles three operator-reliability guardrails (Vikunja #28/#29/#30) plus the
in-flight branch WIP they were built on top of (reasoning_content capture,
operator/project profile editor, write-jail workspaceRoot fix) — the tree is
interdependent (SessionOrchestrator references reasoningArtifactId from the WIP)
and does not compile as separable subsets, so it lands as one commit.

Guardrails:
- #28 mid-stage steering: ClientMessage.SteerSession -> GlobalStreamHandler ->
  orchestrator.submitSteering, reusing SteeringNoteAddedEvent + existing context
  fold (advisory, non-authoritative; invariants #3/#7). Closes the gap where
  steering typed off an approval gate was silently dropped.
- #29 shell-in-file guardrail: ShellInFileContentRule (core:toolintent) blocks a
  file_write whose content is a bare shell command (e.g. "mkdir -p ..."); FileWriteTool
  description now advertises auto-mkdir of parent dirs. Basename-allowlist so the
  extensionless case is caught; scripts/Makefiles/multiline exempt.
- #30 pt1 capability-gap detector: deterministic CapabilityGapDetector maps stage
  intent -> implied ToolCapability, compares to granted tools, emits advisory
  CapabilityGapDetectedEvent in FreestyleDriver.lockAndRun. Recorded, never fails
  the gate and never auto-grants (invariants #3/#4/#5). Reflection rung is pt2.

Verified: ./gradlew check green (whole tree).
This commit is contained in:
2026-07-07 13:27:59 +04:00
parent 879672a47d
commit 41ed6414c6
43 changed files with 1592 additions and 94 deletions
+19 -1
View File
@@ -1,5 +1,16 @@
id = "freestyle_planning"
start = "analyst"
start = "discovery"
# discovery runs before the analyst: it vets the request against the real repo and either clears
# it to plan or parks on a batched list of operator questions (grounds contradictions too, e.g. a
# request naming an endpoint the server doesn't expose). Read-only; owns clarification exclusively.
[[stages]]
id = "discovery"
prompt = "prompts/discovery.md"
produces = [{ name = "discovery", kind = "discovery" }]
allowed_tools = ["file_read", "list_dir", "shell"]
token_budget = 16384
max_retries = 2
# analyst writes no files, but it owns task framing: task_search/task_context (read-only) find
# existing work; task_create (T2, approval-gated — a task is an event-log entry, not a file write)
@@ -24,6 +35,13 @@ produces = [{ name = "execution_plan", kind = "execution_plan" }]
token_budget = 16384
max_retries = 2
[[transitions]]
id = "discovery-to-analyst"
from = "discovery"
to = "analyst"
condition_type = "artifact_validated"
condition_artifact_id = "discovery"
[[transitions]]
id = "analyst-to-architect"
from = "analyst"