v3.1.1 — the pixels match the name now

Brand-asset follow-up to the v3.1.0 rename: the files were renamed then; now the pixels match.

Changed

• Regenerated assets/build-it-header.png, .gif and build-it-animation.mp4 with the build-it wordmark, the "plan-it plans it, build-it builds it." tagline, and /plan-it → /build-it → /review-it lifecycle chips replacing the retired Opus→Fable tiles (GIF 26MB→12.6MB, mp4 3.8MB→0.7MB).
• Regenerated assets/build-it-flow.png with the /build-it pill (same layout).
• Marketplace now self-hosts as build-it — install with /plugin marketplace add DevOtts/build-it then /plugin install build-it@build-it.

Removed

• build-it-field-guide.pdf — orphaned and written around the retired v1 positioning. Git history keeps it.

🤖 Generated with Claude Code

https://claude.ai/code/session_01PozHvVStzWDX3kkHFsMpxr

v3.1.0 — fable-it is now build-it

fable-it is now build-it. What started as a single plugin ("make your model behave like Fable") grew into the build stage of the *-it lifecycle family — plan-it plans it, build-it builds it, review-it reviews it. The name now says what it does. Gates, run-state contract, and verifier protocol are unchanged.

Changed

• Plugin, skill, marketplace entry, and repository renamed fable-it → build-it. GitHub redirects from DevOtts/fable-it remain in place.
• Report folder convention is now .build-it-reports/; a legacy .fable-it-reports/lessons.md is still read for cross-run memory.
• Hardened-mode hooks disable switch is now BUILD_IT_HOOKS_DISABLED=1 (legacy env var still honored).

Kept for back-compat

• /fable-it, "fable it", "fable-it" remain trigger phrases; fable-it stays in keywords — existing prompts keep routing here.
• Historical artifacts (past reports, research docs, earlier changelog entries) keep the old name: they are records, not docs.

If you added the marketplace as DevOtts/fable-it, it keeps working via redirect, but re-adding as DevOtts/build-it is recommended.

fable-it v3.0.1 — safe parallel execution

fable-it v3.0.1 — Safe parallel execution, now proven under contention

This release rolls up v3.0.0 (safe-parallel protocol) and v3.0.1 (interlock hardening) into one deploy.

What's new

Three gates that make parallel and multi-session runs safe by construction (full protocol in parallel-safety.md):

• G-INTERLOCK — atomic run lock. A .taskstate/RUNLOCK acquired with exclusive-create means two fable-it runs can never co-mutate the same tree. The loser blocks with a clear message and resumes cleanly when the lock frees.
• G-WORKTREE — per-lane isolation. Every mutating parallel subagent gets its own git worktree on its own branch. Lanes can't stomp each other's files, and a crashed lane never leaves the main tree dirty.
• G-INTEGRATE — merged-tree integration gate. A slice that's green in isolation but breaks the merged build is reopened, not accepted. Integration checks run on the combined tree before anything is declared done.

Hardening (v3.0.1, CONTRACT v1.1 amendment)

Fixes two protocol races found in the post-release Fable 5 review — gates unchanged:

• TOCTOU-free RUNLOCK acquire — exclusive-create instead of check-then-write.
• Timer-based heartbeat — the lock stays fresh even during long tool calls.
• Dead-owner-only reclaim — a stale lock is reclaimed only when the owning process is provably gone.
• Crash-leftover lane cleanup — orphaned lane branches/worktrees from a previous session are detected and cleaned with a logged note.

Validation

• 6 goldens (T30–T35), each judged by a fresh-context verifier with a citation-lint over transcripts.
• Two A/B rounds vs v2.1.0: identical deliverable quality on a single-lane run, and a 60/46 rubric win on an adversarial parallel run — v2's failure mode (confident false completion with leftover worktrees) is structurally eliminated.
• The interlock was exercised live: two concurrent fable-it sessions on this repo, with the loser blocking and resuming cleanly.

Docs

• New spec: docs/04-v3-safe-parallel-spec.md
• Delivery contract, goldens, and verification records under delivery/v3/

Full changelog: CHANGELOG.md

v2.0.0 — Make your model behave like Fable

From postures to gates. v2 rebuilds fable-it from firsthand Fable 5 research (live harness introspection + Anthropic's "Prompting Claude Fable 5" guide), shipped against a 26-case binding test contract — tabletop goldens judged by fresh-context agents, scripted lints, hook unit tests, and live [REAL] runs on Sonnet 5 and Opus 4.8.

Highlights

• 5-gate catalog replaces posture prose: turn-end, claim, state-change, phase-boundary, delegation — each a trigger + test + action
• Evidence ledger — VERIFIED is a lookup against .taskstate/evidence.md, not a vibe (the Anthropic-measured claim-grounding rule)
• Fresh-context verifier audits every report before it ships (reads only the DoD, the report, and the ledger)
• Disk-backed run state that survives compaction, re-read at every phase boundary + cross-run lessons
• Model-adaptive posture for Sonnet 5 and Opus 4.8, applied at Step 0
• Cost-aware delegation routing — packets routed to model tiers by task shape, spend disclosed per agent
• /launch unattended mode — overnight runs never stall on an approval prompt (human runs keep their gates)
• Hardened mode (opt-in, fail-open) — hooks that mechanically block promise-endings and evidence-free VERIFIED rows
• CDP core single-sourced + route guard — no autonomous write path on authenticated sessions, ever
• Portable root SKILL.md — full mechanics (not descriptions) for Cursor, Codex, Copilot and 70+ agents

Full details: CHANGELOG.md · per-epic board: delivery/STATUS.md · research: docs/research

Install / upgrade

/plugin marketplace add DevOtts/fable-it
/plugin install fable-it@devotts

Other agents: npx skills add DevOtts/fable-it -a <agent>

🤖 Generated with Claude Code

https://claude.ai/code/session_01W61mxW6L7gF9sSgPSUoP6H

fable-it v0.1.0

First tagged release of the fable-it autonomous delivery orchestrator for Claude Code.

• Root SKILL.md manifest (name, description, version, platforms, license) for external skill-marketplace listing.
• Bundled skills: fable-it (conductor), launch, iterate, full-qa, chrome-cdp-control.
• MIT licensed.

Install (Claude Code): /plugin marketplace add DevOtts/fable-it then /plugin install fable-it@devotts.
Install (skills CLI): npx skills add DevOtts/fable-it.

Attached zip artifact is provided for registries (e.g. TrustedSkills) that require a release asset.