issues Search Results · language:Edge language:Python linked:pr language:JavaScript language:Java language:JavaScript
Filter by
5.9M results
Tracks B2.11.2 from longplan_B2_human_governance_web_ui.md.
Goal: prove the web Workbench/server cannot bypass Cosheaf governance.
Required negative tests:
- accepted direct write from artifact editor ...
Summary
After a pipeline run finishes (or halts), there is no structured way to review the full story of what happened — stage
outcomes, timing, blockers, artifacts — in one place. devteam log and devteam ...
enhancement
Summary
Add a customisation / calibration mode that lets a developer interactively fine-tune the positioning of avatar body
parts for each individual outfit, then export those adjustments to a file that ...
Goal
Implement the original M-AUTO-5 milestone: add notification and review escalation for the local A3 workflow.
This should be a local, one-shot operator helper and documentation slice. It should help ...
agent:implement
planning
Add §9.5 to CAPACITY_CONSUMERS_AND_PERCEPTION_SPEC.md documenting the shipped remediation augments (cyber eyes/ears/leg
→ sight/hearing/moving via capacity-bearing replacement organs; voice modulator → ...
documentation
Parent
#10 (Epic: v1.0.0 — Production / adoption bar)
What to build
Investigate whether SessionStart routing-policy injection actually varies across Claude Code surfaces (CLI / IDE /
Claude.ai), and ...
ready-for-agent
Parent
#10 (Epic: v1.0.0 — Production / adoption bar)
Module
Task-class registry — a single source of truth for the canonical task-class list and each class s default tier. One
reason to change: the ...
ready-for-agent
Parent
#10 (Epic: v1.0.0 — Production / adoption bar)
What to build
Extend check_consistency.py to assert that each agent s frontmatter tools: declaration matches its intended capability
class, so a ...
ready-for-agent
Parent
#10 (Epic: v1.0.0 — Production / adoption bar)
What to build
Start cutting semver git tags at each fork release so there s a stable ref to pin, roll back to, and reason about. Tag
the current ...
ready-for-agent
Problem
Screex s value claim is the index is cheaper to feed an LLM than raw frames, at near-equal accuracy. Right now only cost
is measured (scripts/eval.py estimates tokens) — there s no accuracy measurement ...
enhancement

Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.