w00fx/spec-anchored-agentic-development
Agent
Fresh-context, decorrelated reviewer for the implement-feature and implement-backlog loops — context separation, not fresh-context proof: the same model family, spec, and harness share blind spots. Bash is granted for read/verify commands (test runners may write caches and build output). The invariant is non-authoring…
w00fx/spec-anchored-agentic-development
Command
After implementing, generate a complete reference walkthrough of the changes made — every dimension (what, how, why), anchored in the spec, plan, and diff — as a self-contained interactive HTML page in docs/walkthroughs/.
w00fx/spec-anchored-agentic-development
Command
Implement a feature or increment locally, human-driven (implement-feature).
w00fx/spec-anchored-agentic-development
Command
Orchestrate a whole project (or slice) of GitHub issues into resolution-gated waves of Orca child worktrees — one autonomous worker per ticket — and babysit every PR through CI, the applicable-lens review, and third-party feedback until the human merges. Waves are resolution-gated — merged PRs, corroborated no-change…
w00fx/spec-anchored-agentic-development
Command
Generates an implementation plan from a GitHub issue.
w00fx/spec-anchored-agentic-development
Command
Prepare a repository's verification infrastructure for this system — the three-command interface (check / check- / golden), the metric-class gates (stack-agnostic — the class is the requirement, the tool is an instance), the golden harness skeleton, minimum CI, and the ratchet baseline. Brownfield-safe: detect…
w00fx/spec-anchored-agentic-development
Command
Checks drift between specs and implementation.
w00fx/spec-anchored-agentic-development
Command
Run the fresh-context reviewer on a diff, branch, PR, or path (report-only; context separation, not independent proof).
w00fx/spec-anchored-agentic-development
Command
Shape work through a relentless interview — interrogate an idea, a transcript, existing code, or an existing spec until shared understanding; or sharpen a task until it's implementable. Works the question frontier in rounds — every currently-askable question at once, numbered, each with a recommended answer — the…
w00fx/spec-anchored-agentic-development
Command
Break a capability spec (or the shaping conversation that produced it, or a parent issue) into tracer-bullet tickets — vertical slices anchored on the spec's pointed stable IDs (typed, never renumbered), each declaring its blocking edges. Quizzes the human on the breakdown, then publishes to a local tickets.md or to…
w00fx/spec-anchored-agentic-development
Command
Write or update the capability spec from an interview — turns a /shape session (or provided notes/transcript) into specs/ / .md on the template. Does not interview back; gaps become open questions (OQ- -###), never guesses. Every BR/AC carries a stable typed ID (BR- -###, AC- -###) issued in continuation; existing IDs…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
Use when the reviewer subagent is reviewing a diff that implements work defined by a capability spec or an approved plan (protocol Phase 3). Checks two things: does the diff do what the spec requires, and does the diff do what the approved plan said it would. Catches silent drift between intent and implementation.…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
Use when the reviewer subagent is reviewing a diff that touches a domain rule, a calculation, a sensitive numeric value, the audit trail, source-of-rule attribution, stage/responsibility boundaries, or past-period rules. Checks the diff against the system constitution (architecture/constitution.md). This is the…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
Use when the reviewer subagent is reviewing a diff for correctness/logic, simplicity/reuse, test quality, or type design — the default reviewer dimension for the implement-feature and implement-backlog loops. Domain-rule, audit-trail, and spec-conformance concerns are NOT here — route those to…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
EXPLICIT INVOCATION ONLY — run via /implement-backlog or when the user names this skill. Unattended adapter of the shared implementation protocol: autonomous loop over increment issues (bugfix, tech-debt, rule adjustment, reference-data update proposals, feature-within-scope). New capabilities and large features are…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
EXPLICIT INVOCATION ONLY — run via /implement-feature or when the user names this skill. Never auto-trigger on general coding requests. Supervised-local adapter of the shared implementation protocol: code work from existing ratified specs, issues, or backlog items, with the human in the session satisfying every gate.
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
EXPLICIT INVOCATION ONLY — the orchestrated-worker adapter of the shared implementation protocol. Invoked as the FIRST MESSAGE of an Orca child worker: /implement-orchestrated issue # --mode assisted|autonomous. One ticket, one worktree, one PR. Never auto-trigger; never use for local supervised work.
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
Reviews implementation plans before any code is written. Use when the reviewer subagent is judging whether a plan's approach/architecture is sound — approving it or sending it back with approach-level concerns. Used by implement-backlog (Phase 2, as the gate that replaces the human) and available to implement-feature…
w00fx/spec-anchored-agentic-development
Skill Claude CodeCodex
Review tickets (GitHub issues) as executable contracts before dispatch — self-containedness, criteria pointed at the spec, explicit scope and dependencies, size, test scenarios, risk notes. Report-only, [BLOCKER]/[SHOULD]/[NIT]. Use on a generated batch before the quiz (/spec-to-tickets) and before scheduling waves…