codex/skills/accretive-implementer/SKILL.md
Implement, adapt, harden, or repair non-trivial code in a narrow, reviewable, contract-first, witness-backed way. Trigger for planned features, new code, design/plan implementation, migrations, correctness-sensitive refactors, review fixes, bugs, regressions, failing tests, or single-change hardening. Accept review-adjudication/fixed-point-driver handoffs unless stale or contradictory. Do not use for trivial formatting, rote renames, or informational questions.
npx skillsauth add tkersey/dotfiles accretive-implementerInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this as the general coding skill for both implementation and remediation.
Assume the user may only see the last screenful of terminal output.
Operate in UNSOUND, WITNESS-BEARING, PRESERVATION-AWARE, PROGRESS-AWARE, TOTAL, REFINEMENT-FIRST, CONTRACT-FIRST, INVARIANT-FIRST, MECHANISTIC, ACCRETIVE, TRACEABLE, CANONICAL, and SEAM-DISCIPLINED mode.
This skill should extract frontier-model value by turning doctrine into implementation artifacts, not by adding impressive labels.
accretive must cash out as Chosen Cut: stable boundary, why not smaller, why not larger, and proof signal.invariant must cash out as a named invariant plus the owner that enforces or generates it.witness must cash out as a test, command, diff, type/refinement boundary, constructor discipline, or direct artifact check.canonical must cash out as one chosen representation/path and the rejected shadow path when relevant.unwitnessed guarantee and illegal inhabitant must be checked before finalizing non-trivial changes.Before the final Execution Bottom Line, ensure the implementation can answer:
Governing invariant / truth unit:
Canonical owner or chosen cut:
Witness / proof signal:
What invalid state or overbroad behavior remains impossible:
If both are present, start in remediation mode for the broken path, then continue in implementation mode only for the smallest follow-on change required to realize the requested outcome.
Accept handoff from:
Act On, Need Evidence, Handoff Agenda, PR Why Ledger, Governing Invariant Candidate.Rules:
Before editing code on a non-trivial task, determine internally:
Surface these sections when the task is non-trivial, the seam is non-obvious, or upstream evidence is contested.
Use concise sections in this order:
Use concise sections in this order:
tools
Convert markdown plans into beads with dependencies using br CLI. Use when creating task graphs, polishing beads before implementation, or bridging planning to agent swarm execution.
development
Orchestrate Codex skill optimization during active sessions through $cas goal control, $shadow single-session evidence, $tune diagnosis/refinement briefs, and the skill-optimizer custom subagent. Trigger for $opt, skill optimization loops, session-driven skill tuning, meta-skill audits, or explicit validated skill edits. Do not use for general code optimization, product optimization, or performance tuning.
development
Run a targeted fresh-eyes blunder pass over code, specs, plans, adjudications, closure gates, skill edits, or negative-evidence ledgers. Trigger when asked to reread with fresh eyes, find obvious bugs, catch mistakes/oversights/omissions, check for embarrassing misses, or perform a second independent blunder pass before closure. Do not use as a substitute for implementation, adjudication, or verification; use it as the final falsification/check pass for those workflows.
development
Explicitly shadow, tail, watch, follow, monitor, supervise, or companion exactly one Codex session id/path through `$seq`, then apply a named target skill as an interpretation/reporting/proposal/action lens until the watched session stops.