agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Plan an evidence-bound civic newsroom workflow with provenance, privacy, citation, correction, accessibility, and human publication gates.
$ npx -y skills add jmagly/aiwg --skill civic-newsroom-plan --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/civic-newsroom-planContext preview
The summary Claude sees to decide when to auto-load this skill.
Plan an evidence-bound civic newsroom workflow with provenance, privacy, citation, correction, accessibility, and human publication gates.
namespace: aiwg platforms: [all] name: civic-newsroom-plan description: Plan an evidence-bound civic newsroom workflow with provenance, privacy, citation, correction, accessibility, and human publication gates. triggers: - organize a lawful civic newsroom workflow - plan a cited local public information project - build a civic evidence and publishing plan
1. Define the public-interest question, audience, jurisdiction, risk tier, and prohibited uses. 2. Inventory sources and create source/retrieval records; do not acquire before the source gate passes. 3. Map material claims to selectors and epistemic states. 4. Plan accessibility, privacy minimization, response, correction, retention, and named reviewer responsibilities. 5. Use `flows/civic-newsroom.yaml`; stop at its publication gate.
Return an artifact plan, dependency readiness table, open questions, and human decisions. Missing optional frameworks are explicit degraded states.
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