agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Build a vendor-neutral, source-linked public-technology procurement evidence and risk review without autonomous scoring or award recommendations.
$ npx -y skills add jmagly/aiwg --skill public-technology-review --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/public-technology-reviewContext preview
The summary Claude sees to decide when to auto-load this skill.
Build a vendor-neutral, source-linked public-technology procurement evidence and risk review without autonomous scoring or award recommendations.
namespace: aiwg platforms: [all] name: public-technology-review description: Build a vendor-neutral, source-linked public-technology procurement evidence and risk review without autonomous scoring or award recommendations. triggers: - review a public technology procurement - compare civic technology contract evidence - assess public sector vendor lock in and lifecycle risk
1. Select and cite the governing procurement profile; FAR/NIST are reference models unless adopted by that jurisdiction. 2. Inventory need, planning, market research, solicitation, selection, award, administration, oversight, and public-release records. 3. Separate protected proposals/source-selection material from public releases. 4. Record claims and evidence without sentiment scores. Risk ratings require a human-defined scale, rationale, uncertainty, owner, and verification method. 5. Do not infer corruption, rank vendors, or recommend an award autonomously. 6. Record explicit handoff states for missing records, meeting/vote analysis, and any publication packet; protected material never enters a public handoff.
A schema-valid evidence/risk review and human decision questions.
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