agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Assess workspace alignment and recommend cleanup or realignment actions at key lifecycle transition points
$ npx -y skills add jmagly/aiwg --skill workspace-health --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/workspace-healthContext preview
The summary Claude sees to decide when to auto-load this skill.
Assess workspace alignment and recommend cleanup or realignment actions at key lifecycle transition points
namespace: aiwg name: workspace-health platforms: [all] description: Assess workspace alignment and recommend cleanup or realignment actions at key lifecycle transition points
<!-- AIWG-SKILL-CALLOUT --> > **Skill access pattern (post-kernel-pivot, 2026.5+)** > > Skill names referenced in this document are AIWG skills, **not slash commands**. Most are not kernel-listed and cannot be invoked as `/skill-name` by the platform. Reach them via: > > ```bash > aiwg discover "<capability>" > aiwg show skill <name> > ``` > > Only kernel-listed skills (`aiwg-doctor`, `aiwg-refresh`, `aiwg-status`, `aiwg-help`, `use`, `steward`) are directly invokable as slash commands. See [skill-discovery rule](../../../addons/aiwg-utils/rules/skill-discovery.md).
Assesses workspace alignment and suggests cleanup or realignment actions at key transition points.
> As of ADR-021, `workspace-health` delegates structural checks to the semantic memory kernel.
**Delegation pattern**: 1. `workspace-health` retains its consumer-neutral health-check UX 2. Runs `memory-lint` for every installed framework in `.aiwg/frameworks/registry.json` 3. Aggregates results across all consumers into a unified report 4. `aiwg doctor` continues to call this skill unchanged
**Backward compatibility**: No UX changes. Output format unchanged.
@agentic/code/addons/semantic-memory/skills/memory-lint/SKILL.md
Alternate expressions and non-obvious activations (primary phrases are matched automatically from the skill description):
Auto-triggers:
This skill is commonly invoked:
checks:
- name: working_directory_size
description: Check if .aiwg/working/ has accumulated stale files
threshold: ">10 files or >1MB"
action: Suggest /workspace-prune-working
- name: orphan_drafts
description: Draft artifacts not linked to requirements
action: Suggest review or archival
- name: stale_locks
description: Lock files older than 24h
action: Suggest cleanupchecks:
- name: phase_documentation
description: Current phase docs match project state
sources:
- .aiwg/planning/phase-plan-*.md
- .aiwg/reports/*-completion-report.md
action: Suggest /workspace-realign if mismatched
- name: requirement_coverage
description: All requirements have linked artifacts
action: Suggest /check-traceability
- name: architecture_drift
description: Code diverged from documented architecture
action: Suggest architecture review or ADR updatechecks:
- name: stale_artifacts
description: Key artifacts not updated in >30 days during active dev
artifacts:
- SAD (Software Architecture Document)
- Risk Register
- Test Strategy
action: Flag for review
- name: completion_markers
description: Artifacts marked complete but phase still active
action: Suggest status update## Workspace Health Report **Overall Status**: [Healthy | Needs Attention | Requires Realignment] ### Quick Actions - [ ] Run `workspace-prune-working` - 15 stale files in working/ - [ ] Review 3 orphaned draft artifacts - [ ] Update risk register (last modified 45 days ago) ### Detailed Findings #### Working Directory - Status: Needs cleanup - Files: 15 (threshold: 10) - Oldest: inception-notes-draft.md (created 2024-11-15) - Recommendation: Promote or archive before next phase #### Documentation Alignment - Phase: Construction - Last phase report: Elaboration completion (2024-12-01) - Missing: Construction kickoff documentation - Recommendation: Run `flow-elaboration-to-construction` completion steps #### Traceability - Requirements covered: 85% - Orphan code files: 3 - Recommendation: Run `check-traceability` for details
Add to flow command templates:
## Post-Completion After this flow completes, consider running a workspace health check: [workspace-health] Assessing workspace alignment... If issues found, the skill will suggest appropriate cleanup commands.
The orchestrator should invoke this skill: 1. When transitioning between SDLC phases 2. After completing iteration cycles 3. When user requests project status 4. Before major deployments
This skill should: 1. Read workspace state from `.aiwg/` structure 2. Compare against expected state for current phase 3. Generate actionable recommendations 4. NOT automatically execute cleanup (user confirms)
Reusable project context and specialist workflows for the AI tools you already use. Plan software, coordinate specialist reviews, prepare campaigns, investigate incidents, organize research, curate media, and maintain operational knowledge.
Repo: jmagly/aiwg
Crash-resilient external agent loop with state persistence and CI/CD integration
Detect requests for iterative autonomous agent loops and route to the appropriate loop executor
Automatically execute tests when code-generating agents modify source files, enforcing the execute-before-return pattern
Enable agent loops to learn from similar past tasks and share patterns across loops
Query and manage the executable feedback debug memory
Execute tests on generated code and iterate until passing