jeremy
Loading activity
Derp in the swamp
- On AImerelinux.orgapproved
- Weirtwonines.codeberg.pageapproved
- What the Factory Forgetstwonines.codeberg.pageapproved
- Building an Adversarial Knowledge Base for AI Agentstwonines.codeberg.pageapproved
- #1795Workflow summary omits failed assertion stepsShippedbug
- #1775Add typed read-only remote ref lookup to @swamp/gitShippedfeature
- #1739@swamp/git: no way to amend a commit, but push can --forceShippedfeature
- #1708@swamp/git: add a read-only working-tree diff for pre-commit reviewShippedfeature
- #1705workflow approvals displays the latest run's prompt for every suspended runShippedbug
Github Pr
Narrow GitHub.com same-repository pull-request boundary: preflight, plan, apply, and read-back verification. Delivery supplies a body that exactly matches the commit body; this adapter preserves it and computes change integrity and readiness internally. No merge, release, repository administration, PR update, or arbitrary endpoint capability.
Mere Pm Dev
Build and validate the Mere package-manager source tree through fixed Zig commands run inside a @jeremy/mere-shell namespace, so no host Zig is needed, and package its fresh-branch delivery workflows with commit bodies that are copied exactly into pull-request descriptions and explicit issue-closing actions. Does not build recipes, publish artifacts, or provide credentials.
Mere Dev
Everything for MereLinux recipes: scan the tree and look up upstream versions, bump or rebuild a recipe, and validate, build, and import it with mere, then deliver the change. Pull-request bodies are derived from commit messages and must match them exactly. Invokes mere directly without a shell wrapper; installed-package smoke tests run in @jeremy/mere-shell.
Mere Shell
Execute typed argv inside a verified, cancellable Mere Linux shell namespace with dedicated roots and isolated package sets.
Nmap
Network scanning with nmap, stored as structured data for change tracking over time. Produces host and port resources keyed by IP and port number, so consecutive scans naturally version and diff via the swamp data model.