agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Submit a bug report, feature request, or feedback to the AIWG GitHub repository — prefills system context automatically
$ npx -y skills add jmagly/aiwg --skill feedback --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/feedbackContext preview
The summary Claude sees to decide when to auto-load this skill.
Submit a bug report, feature request, or feedback to the AIWG GitHub repository — prefills system context automatically
namespace: aiwg name: feedback platforms: [all] description: Submit a bug report, feature request, or feedback to the AIWG GitHub repository — prefills system context automatically
Submit a bug report, feature request, doc gap, or general feedback to the AIWG GitHub repository. System context (version, OS, provider, installed frameworks) is collected and prefilled automatically.
Alternate expressions and non-obvious activations (primary phrases are matched automatically from the skill description):
| Pattern | Example | Action | |---------|---------|--------| | Bug report | "report a bug with aiwg serve" | `aiwg feedback --type bug` | | Feature request | "request a feature for aiwg" | `aiwg feedback --type feature` | | Doc gap | "the docs for mcp inject are wrong" | `aiwg feedback --type doc` | | General | "I have feedback" | `aiwg feedback` | | With context | "file this as a bug: X" | Extract title/body, run `aiwg feedback --type bug --title "..." --body "..."` |
When triggered:
1. **Extract from conversation** (if the user described the issue):
2. **Run the appropriate command**:
# Interactive (when type/title/body not clear from context) aiwg feedback # With extracted type aiwg feedback --type bug # Fully extracted from conversation aiwg feedback --type bug --title "doctor crashes when .aiwg missing" --body "Running aiwg doctor in a new project with no .aiwg directory causes an unhandled exception..." # Feature request aiwg feedback --type feature --title "add --watch flag to aiwg index build" --body "..." # Doc gap aiwg feedback --type doc --title "mcp inject workflow not documented" --body "..."
3. **Report the result** — confirm the issue was filed or provide the GitHub URL.
**User**: "aiwg doctor crashes when I run it in a new project — there's an unhandled error about missing .aiwg"
**Extraction**:
**Action**:
aiwg feedback --type bug \ --title "doctor crashes when .aiwg directory is missing" \ --body "Running aiwg doctor in a project with no .aiwg directory causes an unhandled exception. Steps: 1) Create a new empty directory 2) Run aiwg doctor 3) Error occurs."
**User**: "I wish aiwg session could pass extra flags directly to the provider binary"
**Extraction**:
**Action**:
aiwg feedback --type feature \ --title "aiwg session: pass-through flags to provider binary" \ --body "Would be useful to be able to pass provider-specific flags through aiwg session, e.g. aiwg session -- --verbose"
**User**: "I have some feedback"
**Clarification prompt**: "What's the feedback about? (Bug you found, feature you'd like, or something else?)"
Then run interactively:
aiwg feedback
If the user's intent is ambiguous:
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