agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Reconcile a supplied public-meeting transcript, vote ledger, and official minutes while preserving uncertainty and requiring human verification.
$ npx -y skills add jmagly/aiwg --skill public-meeting-reconcile --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/public-meeting-reconcileContext preview
The summary Claude sees to decide when to auto-load this skill.
Reconcile a supplied public-meeting transcript, vote ledger, and official minutes while preserving uncertainty and requiring human verification.
namespace: aiwg platforms: [all] name: public-meeting-reconcile description: Reconcile a supplied public-meeting transcript, vote ledger, and official minutes while preserving uncertainty and requiring human verification. script: entrypoint: scripts/meeting_reconcile.mjs runtime: node cwd: project-root argsHint: "<vote-ledger.json> <meeting-reconciliation.json>" triggers: - reconcile a public meeting transcript with minutes - validate a vote ledger - compare meeting votes and approved minutes
1. Resolve the jurisdiction/source/recording posture before media acquisition. 2. Reuse media-curator transcription and research-complete media induction when installed; otherwise emit `blocked-dependency-missing` rather than text. 3. Keep media, transcript, vote ledger, and minutes reconciliation versioned separately. Record agenda item, motion text/state, mover, seconder, timestamp, citation, member votes/abstentions/recusals/absences, and both announced and calculated results. Never identify a speaker from diarization alone. 4. Run the executable gate. Conflicts, inferred votes, unverified official labels, and missing source cues block. 5. A human verifies each material motion, vote, speaker label, and public/closed boundary before any handoff to publication.
JSON gate results plus a reviewed reconciliation packet; never official minutes.
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