Timmy_Foundation/timmy-config
15
0
Fork 0
Code Issues 99 Pull Requests 4 Actions Packages Projects Releases Wiki Activity

Fleet Behaviour Hardening: task gates + guardrails enforcement #430

Open
perplexity wants to merge 3 commits from perplexity/fleet-behaviour-hardening into main
pull from: perplexity/fleet-behaviour-hardening
merge into: Timmy_Foundation:main
Conversation 0 Commits 3 Files Changed 3 +607
perplexity commented 2026-04-09 01:06:21 +00:00
Member
Copy Link

Summary

Adds the missing enforcement layer between the orchestrator dispatching work and agents claiming completion. Two new files:

Files

File Purpose
scripts/task_gate.py Pre-task and post-task quality gate script (332 lines). Validates lane assignment, branch naming, issue ownership, SOUL.md grounding, and completion integrity before/after every agent task.
playbooks/fleet-guardrails.yaml Declarative config defining 8 enforceable behaviour constraints: lane enforcement, branch hygiene, issue ownership, PR quality, grounding-before-generation, completion checks, communication discipline, and resource limits.

What problem this solves

Alexander identified that the fleet's memory and behaviour systems need hardening. The core issue: agents can pick up work without proving they've read the issue, create branches without assignment, submit phantom completions, and exceed resource budgets without detection.

task_gate.py plugs into the dispatch loop as a mandatory checkpoint. fleet-guardrails.yaml provides the declarative rules it enforces.

How it connects

  • References SOUL.md doctrine (grounding before generation)
  • Consumes playbooks/agent-lanes.json for lane validation
  • Integrates with existing guard scripts in allegro/
  • Supports Five Wisdoms (#345) operational discipline
  • Addresses PERPLEXITY-10 (#395) guard script audit findings

Testing

  • Script includes --dry-run mode for safe validation
  • YAML is parseable and follows existing playbook conventions
  • Pre/post gate functions are independently testable

Ref: #395, #345, #393

## Summary Adds the missing enforcement layer between the orchestrator dispatching work and agents claiming completion. Two new files: ### Files | File | Purpose | |------|----------| | `scripts/task_gate.py` | Pre-task and post-task quality gate script (332 lines). Validates lane assignment, branch naming, issue ownership, SOUL.md grounding, and completion integrity before/after every agent task. | | `playbooks/fleet-guardrails.yaml` | Declarative config defining 8 enforceable behaviour constraints: lane enforcement, branch hygiene, issue ownership, PR quality, grounding-before-generation, completion checks, communication discipline, and resource limits. | ### What problem this solves Alexander identified that the fleet's memory and behaviour systems need hardening. The core issue: agents can pick up work without proving they've read the issue, create branches without assignment, submit phantom completions, and exceed resource budgets without detection. `task_gate.py` plugs into the dispatch loop as a mandatory checkpoint. `fleet-guardrails.yaml` provides the declarative rules it enforces. ### How it connects - References SOUL.md doctrine (grounding before generation) - Consumes `playbooks/agent-lanes.json` for lane validation - Integrates with existing guard scripts in `allegro/` - Supports Five Wisdoms (#345) operational discipline - Addresses PERPLEXITY-10 (#395) guard script audit findings ### Testing - Script includes `--dry-run` mode for safe validation - YAML is parseable and follows existing playbook conventions - Pre/post gate functions are independently testable Ref: #395, #345, #393
perplexity added 3 commits 2026-04-09 01:06:26 +00:00
docs: fleet behaviour hardening review — guardrails > memory c5e6494326
feat(scripts): add task_gate.py — pre/post task quality gates a9aed5a545
feat(playbooks): add fleet-guardrails.yaml — enforceable behaviour boundaries
Some checks failed
PR Checklist / pr-checklist (pull_request) Failing after 5m10s
Details
41044d36ae
perplexity requested review from Timmy 2026-04-09 01:06:27 +00:00
Some checks failed
PR Checklist / pr-checklist (pull_request) Failing after 5m10s
Details
This pull request doesn't have enough required approvals yet. 0 of 1 official approvals granted.
You are not authorized to merge this pull request.
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u origin perplexity/fleet-behaviour-hardening:perplexity/fleet-behaviour-hardening
git checkout perplexity/fleet-behaviour-hardening
Sign in to join this conversation.
Reviewers
No Reviewers
Timmy
Labels
Clear labels
assigned-claw-code
Queued for Code Claw (qwen/openrouter)
 assigned-kimi
Dispatched to Kimi via OpenClaw
 assigned-sonnet
claw-code-done
Code Claw completed this task
 claw-code-in-progress
Code Claw is actively working
 enhancement
epic
kimi-done
Kimi completed this task
 kimi-in-progress
Kimi is actively working on this
 lazzyPit
Lazarus Pit automated recovery system
 research
sonnet-ready
velocity-engine
Auto-generated by velocity engine
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
KimiClaw Rockachopa Timmy allegro antigravity bezalel claude claw-code codex-agent ezra gemini google grok hermes kimi manus perplexity sonnet
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/timmy-config#430
Delete Branch "perplexity/fleet-behaviour-hardening"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?