devlyn-cli 4.2.26 → 4.3.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/bin/devlyn.js +510 -0
- package/bin/engines.js +20 -15
- package/bin/harness-activation.js +1 -2
- package/bin/release.js +16 -3
- package/bin/skill-history.json +5 -0
- package/config/skills/_shared/harness-store.py +202 -0
- package/config/skills/_shared/task-completion.md +6 -3
- package/config/skills/_shared/warm-build-dirs.py +97 -2
- package/config/skills/devlyn-ideate/SKILL.md +5 -1
- package/config/skills/devlyn-ideate/references/loop.md +12 -2
- package/config/skills/devlyn-ideate/references/plan-add.md +12 -0
- package/config/skills/devlyn-ideate/scripts/acceptance.py +5 -0
- package/config/skills/devlyn-ideate/scripts/queue.py +165 -15
- package/harness/AGENTS.md +37 -0
- package/harness/CLAUDE.md +37 -0
- package/native/claude/.claude-plugin/plugin.json +6 -0
- package/native/claude/skills/_shared/.ruff.toml +2 -0
- package/native/claude/skills/_shared/adapters/claude.md +15 -0
- package/native/claude/skills/_shared/adapters/codex.md +14 -0
- package/native/claude/skills/_shared/adapters/omp.md +6 -0
- package/native/claude/skills/_shared/codex-monitored.sh +263 -0
- package/native/claude/skills/_shared/engine-doctor.sh +70 -0
- package/native/claude/skills/_shared/engine-preflight.md +29 -0
- package/native/claude/skills/_shared/expected-contract.py +242 -0
- package/native/claude/skills/_shared/expected.schema.json +120 -0
- package/native/claude/skills/_shared/harness-store.py +202 -0
- package/native/claude/skills/_shared/invocation-receipt.py +499 -0
- package/native/claude/skills/_shared/platform-support.py +395 -0
- package/native/claude/skills/_shared/role-config.py +247 -0
- package/native/claude/skills/_shared/runtime-principles.md +35 -0
- package/native/claude/skills/_shared/task-complete.py +2865 -0
- package/native/claude/skills/_shared/task-completion.md +249 -0
- package/native/claude/skills/_shared/warm-build-dirs.py +1065 -0
- package/native/claude/skills/devlyn-engines/SKILL.md +63 -0
- package/native/claude/skills/devlyn-ideate/SKILL.md +132 -0
- package/native/claude/skills/devlyn-ideate/references/elicitation.md +29 -0
- package/native/claude/skills/devlyn-ideate/references/loop.md +123 -0
- package/native/claude/skills/devlyn-ideate/references/package-format.md +128 -0
- package/native/claude/skills/devlyn-ideate/references/plan-add.md +12 -0
- package/native/claude/skills/devlyn-ideate/scripts/acceptance.py +882 -0
- package/native/claude/skills/devlyn-ideate/scripts/queue.py +2191 -0
- package/native/codex/.agents/plugins/marketplace.json +12 -0
- package/native/codex/plugins/devlyn/.codex-plugin/plugin.json +6 -0
- package/native/codex/plugins/devlyn/skills/_shared/.ruff.toml +2 -0
- package/native/codex/plugins/devlyn/skills/_shared/adapters/claude.md +15 -0
- package/native/codex/plugins/devlyn/skills/_shared/adapters/codex.md +14 -0
- package/native/codex/plugins/devlyn/skills/_shared/adapters/omp.md +6 -0
- package/native/codex/plugins/devlyn/skills/_shared/codex-monitored.sh +263 -0
- package/native/codex/plugins/devlyn/skills/_shared/engine-doctor.sh +70 -0
- package/native/codex/plugins/devlyn/skills/_shared/engine-preflight.md +29 -0
- package/native/codex/plugins/devlyn/skills/_shared/expected-contract.py +242 -0
- package/native/codex/plugins/devlyn/skills/_shared/expected.schema.json +120 -0
- package/native/codex/plugins/devlyn/skills/_shared/harness-store.py +202 -0
- package/native/codex/plugins/devlyn/skills/_shared/invocation-receipt.py +499 -0
- package/native/codex/plugins/devlyn/skills/_shared/platform-support.py +395 -0
- package/native/codex/plugins/devlyn/skills/_shared/role-config.py +247 -0
- package/native/codex/plugins/devlyn/skills/_shared/runtime-principles.md +35 -0
- package/native/codex/plugins/devlyn/skills/_shared/task-complete.py +2865 -0
- package/native/codex/plugins/devlyn/skills/_shared/task-completion.md +249 -0
- package/native/codex/plugins/devlyn/skills/_shared/warm-build-dirs.py +1065 -0
- package/native/codex/plugins/devlyn/skills/devlyn-engines/SKILL.md +63 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/SKILL.md +132 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/references/elicitation.md +29 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/references/loop.md +123 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/references/package-format.md +128 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/references/plan-add.md +12 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/scripts/acceptance.py +882 -0
- package/native/codex/plugins/devlyn/skills/devlyn-ideate/scripts/queue.py +2191 -0
- package/package.json +3 -2
- package/release.json +1630 -0
|
@@ -0,0 +1,35 @@
|
|
|
1
|
+
The following Devlyn instructions are defaults for this session. Applicable explicit user instructions and native owner/project instructions take precedence over these defaults. This delegation applies only to the Devlyn defaults below; it does not change other engine instructions or permissions.
|
|
2
|
+
|
|
3
|
+
# Project Instructions
|
|
4
|
+
|
|
5
|
+
devlyn-cli installs the `devlyn-ideate` and `devlyn-engines` skills alongside the principles below, which are non-negotiable on every change — yours and any sub-agent's.
|
|
6
|
+
|
|
7
|
+
## North Star
|
|
8
|
+
|
|
9
|
+
This contract serves one goal: any capable engine — Claude, GPT/Codex, or a future adapter-equipped model — takes a user's intent (prompt, spec, or queue entry) end-to-end to shipped, engineer-quality software, hands-free, with consistent quality across engines. When rules below conflict or feel ambiguous in context, resolve toward this goal.
|
|
10
|
+
|
|
11
|
+
## Core principles
|
|
12
|
+
|
|
13
|
+
Seven rules govern every change. Cite them by name when a decision touches one.
|
|
14
|
+
|
|
15
|
+
1. **No workaround** — fix the root cause, never the symptom. No `any`, no `@ts-ignore`, no silent `catch`, no hardcoded fallback that hides a broken contract. No config bypasses.
|
|
16
|
+
2. **No overengineering** — smallest change that closes the goal. New abstractions require an observed failure mode they prevent. Subtractive-first: ask "what can I delete instead?" before writing anything new. Remove code your change makes unused; only report unrelated pre-existing dead code.
|
|
17
|
+
3. **No guesswork** — verify with the actual files, logs, diffs, and run output before forming conclusions. State the falsifiable prediction BEFORE the experiment; record raw results AFTER. Retroactive prediction edits are dishonest.
|
|
18
|
+
4. **Worldclass** — code that survives review at a non-trivial codebase. Zero CRITICAL, zero HIGH security/design findings on the shippable path.
|
|
19
|
+
5. **Best practice** — idiomatic for the language and framework. Use standard primitives; do not hand-roll what the library already provides.
|
|
20
|
+
6. **Optimized** — use efficient code and avoid unnecessary work without sacrificing correctness or the user's requirements.
|
|
21
|
+
7. **Production ready** — error states are explicit and visible; behavior under failure is what the user expects, not silent corruption.
|
|
22
|
+
|
|
23
|
+
Three discipline rules govern HOW the principles are applied:
|
|
24
|
+
|
|
25
|
+
- **Root cause via flexible why-chain.** Keep asking "why?" until you find the violated invariant. **If the answer surfaces in 2 questions, stop.** If it takes 5 or 7, keep going. Strict counts are wrong; until-found is right.
|
|
26
|
+
- **First-principles thinking.** Challenge the requirement before optimizing the answer. Surface unstated assumptions, ambiguities, tradeoffs, and simpler alternatives BEFORE implementing — do not silently pick one interpretation when multiple exist, do not hide confusion, push back when a simpler path is genuinely better. Most "we have to do X" assumptions are habit, not necessity. Reduce the problem to its irreducible truths and rebuild from there.
|
|
27
|
+
- **Perfection is achieved not when there is nothing more to add, but when there is nothing left to take away.** — Saint-Exupéry. This is the operating definition of "done." A change is finished when no further line, branch, flag, or doc paragraph can be removed without breaking a learned failure mode. Not before.
|
|
28
|
+
|
|
29
|
+
## Quick Start
|
|
30
|
+
|
|
31
|
+
- `devlyn-ideate` — loop designer and intent queue: `plan <intent or document>` writes a validated loop package, `add` queues its tasks, `status` (or no arguments) reports the queue, `drain` executes it serially.
|
|
32
|
+
- Delivery — direct work edits the current checkout; deliver completed, verified requests with changes by project policy (default PR and merge), following `_shared/task-completion.md` in `.claude/skills/` or `.agents/skills/` (project, else `~`). With known concurrent writers, isolate before editing; on unexplained changes, pause writes until safe or isolated. An ideate drain delivers every task. Explicit limits win; commit alone stays local.
|
|
33
|
+
- Executor — `devlyn-engines` shows it and pins it in `.devlyn/engines.json`. A pinned executor does the implementation work, direct or drained: when it is not you, delegate to it, and when it is unavailable stop with `BLOCKED:<engine>-unavailable`. Without a pin, you are the executor.
|
|
34
|
+
|
|
35
|
+
Each skill's `SKILL.md` is the source of truth for its flags and workflow.
|